Beyond carbon: reactive nitrogen emissions from ammonia as a marine fuel and implications for ocean ecosystems

Abstract

Maritime shipping, responsible for ~3% of global carbon dioxide (CO₂) emissions, must rapidly decarbonise to meet the International Maritime Organization’s target of net-zero greenhouse gas (GHG) emissions by 2050. Renewable-based ammonia has emerged as a leading near-zero-carbon fuel candidate; however, its reactive nitrogen (Nr) footprint remains poorly characterised. Across the ammonia marine fuel (AMF) value chain, Nr emissions arise as nitrogen oxides (NO_x) and nitrous oxide (N₂O) from combustion, and ammonia (NH₃) losses to air (i.e., fugitive leaks and operational emissions) during production, storage, bunkering, and shipboard use, alongside episodic releases to water from spills and nitrogen-bearing effluent. This review synthesises current evidence on Nr emissions across the AMF value chain, examines plausible pathways through which their downstream transport, deposition, and fate may impact marine ecosystems and climate, and highlights key knowledge gaps. We show that resulting impacts—including altered marine productivity and community structure, eutrophication, biodiversity loss, and indirect N₂O formation—are geographically dependent. Identical emissions can produce markedly different outcomes depending on whether deposition and other inputs occur in oligotrophic gyres, coral reefs, oxygen minimum zones, or already-stressed coastal and marginal seas. We also highlight potentially vulnerable regions where major shipping corridors overlap with nitrogen-sensitive marine ecosystems and Nr inputs may generate disproportionate ecological or climate impacts. Finally, we review mitigation measures and governance gaps across the AMF lifecycle. Realizing the full climate and environmental benefits of AMF will depend on integrated lifecycle accounting, stringent emissions controls, and policies that address nitrogen-cycle impacts and protect marine ecosystems alongside carbon reductions.

1 Introduction

International maritime shipping carries over 80% of world trade by volume and accounts for approximately 3% of carbon dioxide (CO₂) emissions, leaving the International Maritime Organization’s (IMO) 2050 net-zero target dependent on near-complete replacement of conventional fossil fuels within two decades (IMO, 2020a, 2023). Among candidate alternative fuels, ammonia (NH₃) produced from renewable electricity (i.e., “green” or “electro [e-]” NH₃; see a brief description of NH₃ production routes in Box 1) has emerged as a frontrunner on cost and energy density grounds. Its contribution to the marine energy mix is projected to reach 6–10% by 2030 and 35–60% by 2050, while some net-zero scenarios envisage a market share of more than 40% by 2050 (IEA, 2023). Pure NH₃ contains no carbon, and its carbon-free combustion has driven its policy attractiveness.

Yet near carbon-free combustion does not imply impact-free deployment. Combustion of NH₃ produces reactive nitrogen (Nr) species, including nitrogen oxides (NO_x), nitrous oxide (N₂O), and unreacted NH₃ slip from incomplete combustion (Bertagni et al., 2023; Schmidt et al., 2025) While N₂O is a potent GHG, with ~273 times the global warming potential (GWP) of CO₂ over 100 years and an ozone-depleting substance, NH₃ and NO_x are short-lived air pollutants that contribute to fine particulate matter (PM_2.5) formation and associated health impacts (Forster et al., 2021). Accordingly, replacing hydrocarbon fuels with NH₃ shifts the burden of emissions from the carbon cycle to the nitrogen cycle, with implications for climate, health and the environment (see Box 2).

Besides combustion, other Nr emissions can occur across the ammonia marine fuel (AMF) value chain regardless of production route. As shown in Figure 1, this value chain and the processes relevant to the fate of related Nr emissions encompass four stages: (1) NH₃ production; (2) storage, transport, and bunkering; (3) shipboard combustion; and (4) atmospheric transport, ocean deposition, and marine biogeochemical transformation of Nr (IEA, 2021; Schmidt et al., 2025). In addition to combustion emissions, Nr emissions across the lifecycle include continuous NH₃ losses to the atmosphere (e.g., fugitive leaks and boil-off gas) and episodic releases to water (accidental spills and nitrogen-bearing effluent), while indirect N₂O can be formed after emitted Nr is transported, deposited, and transformed in natural environments (Esquivel-Elizondo et al., 2025). First-order global analysis indicates that total Nr emissions from AMF could be ~66% lower than current very-low-sulfur fuel oil under stringent controls, or up to ~185% higher without them (Esquivel-Elizondo et al., 2026a). These estimates cover stages 1–3 (the value chain); the fourth stage—transport, deposition, and marine ecosystem fate—is the critical extension beyond traditional three-stage NH₃ inventories and assessments (Galloway et al., 2008; Sutton et al., 2011).

Figure 1

Bertagni et al. (2023).

This downstream fate is especially important in the marine context. Unlike terrestrial NH₃ applications, whose impacts are largely confined to agricultural regions and their immediate drainage basins, onboard handling of AMF and its combustion release NO_x and NH₃ along transoceanic shipping corridors. These emissions can be transported through the atmosphere and deposited elsewhere over ocean regions with very different sensitivities, ranging from nitrogen-replete upwelling zones to severely nitrogen-limited oligotrophic gyres, coral reef systems, oxygen minimum zones (OMZs), and coastal or marginal seas (Corbett and Koehler, 2003; Duce et al., 2008; Naqvi et al., 2010; Moore et al., 2013). As a result, ecological responses can differ markedly across receiving environments.

In coastal and shelf systems, additional atmospheric Nr can initiate eutrophication, in which elevated nutrient inputs stimulate phytoplankton and macroalgal blooms, whose subsequent senescence and microbial decomposition deplete dissolved oxygen in subsurface waters, ultimately displacing mobile fauna and causing mortality in sessile and less tolerant species (Diaz and Rosenberg, 2008; Rabalais et al., 2010). In oxygen-depleted waters, additional atmospheric Nr can also contribute to indirect N₂O formation and associated climate forcing (Suntharalingam et al., 2012).

These linked atmospheric and marine processes are captured by the concept of the “nitrogen cascade,” which provides the conceptual framework linking emissions, transport, deposition, and ecosystem transformation. It describes the sequential transformation of Nr through multiple environmental reservoirs, with a single molecule of anthropogenic Nr potentially generating impacts across atmospheric, terrestrial, aquatic, and marine systems before returning to inert dinitrogen gas (N₂) (Box 2) (Galloway et al., 2003, 2008). For AMF, this framework, together with the lifecycle perspective, highlights two key implications. First, Nr losses are distributed across the value chain rather than concentrated at the engine, so combustion-only emission factors understate total exposure. Second, it decouples emission location from impact location: Nr released along shipping corridors can be transported, deposited, and transformed downstream, causing diverse impacts in different ecosystems, such as eutrophication, acidification, biodiversity loss, and indirect N₂O formation in oxygen-depleted waters (Duce et al., 2008; Naqvi et al., 2010; Suntharalingam et al., 2012). Under weak emissions controls, these indirect N₂O pathways could offset an estimated 15–40% of AMF’s potential climate benefit (Bertagni et al., 2023; Esquivel-Elizondo et al., 2025).

Because agricultural NH₃ production is not expected to decline, AMF deployment would occur alongside continued agricultural NH₃ use. AMF demand could increase global NH₃ production by an additional 3–4 times by mid-century (Figure 2), compounding anthropogenic Nr emissions by adding a new anthropogenic Nr source concentrated around industrial hubs, ports, and shipping corridors (Erisman, 2026). This expansion would occur within a nitrogen cycle that already exceeds the planetary boundary for Nr inputs—the extent of perturbation the global nitrogen cycle can undergo before posing unsafe conditions for humanity (62–82 Tg N yr^-1)—with current anthropogenic Nr emissions estimated at ~210 Tg N yr^-1 (Stevens, 2019; Richardson et al., 2023). Consequently, new Nr sources associated with AMF could intensify existing risks related to climate, air quality, ecosystem degradation, and biodiversity loss (Suntharalingam et al., 2012).

Figure 2

Bertagni et al. (2023); Chen et al. (2024) for sectoral allocation.

Accordingly, whether AMF delivers net climate and environmental benefits depends on lifecycle emissions controls, requiring a comprehensive assessment of Nr emissions and their direct and indirect impacts to inform effective mitigation and regulation (Chen et al., 2024; Esquivel-Elizondo et al., 2026a). Existing regulatory and certification frameworks are largely designed for agricultural systems and direct GHGs, with system boundaries focused on fertiliser production, field volatilization, and soil N₂O emissions (Galloway et al., 2003; Sutton et al., 2011), and thus only partially capture AMF-related Nr risks. AMF introduces additional transport, deposition, and marine biogeochemical pathways that are not represented in agricultural accounting frameworks (Wolfram et al., 2022; Bertagni et al., 2023). A dedicated lifecycle accounting framework is therefore needed to support robust policy and industry decision-making for NH₃ use in the maritime sector.

This review synthesizes evidence on Nr emissions across the AMF value chain (stages 1–3) and their implications for sensitive ocean ecosystems (stage 4) (Figure 1). It examines emission sources, marine fate and ecosystem impacts (Table 1), as well as mitigation options and governance gaps. Non-marine impacts are considered only where needed to interpret lifecycle performance or policy relevance. Organised around the AMF lifecycle and the nitrogen cascade framework, the review assesses the conditions under which AMF can deliver real climate benefits without shifting unacceptable burdens to the nitrogen cycle and marine environment.

1.1 Scope and approach

This article is a narrative review rather than a systematic review. Literature was identified through searches of Scopus and Google Scholar, supplemented by regulatory and technical documents from the IMO, the European Union, national authorities, and the International Energy Agency (IEA). Primary search terms combined fuel- and emissions-related keywords (“ammonia”, “marine fuel”, “maritime alternative fuels”, “maritime decarbonisation”, “NO_x”, “N₂O”, “ammonia slip”, “reactive nitrogen”, “nitrogen cascade”, “nitrogen deposition”) with receiving-environment terms (“ocean denitrification”, “oxygen minimum zone”, “eutrophication”, “coral reef”, “seagrass”, “marginal sea”, “oligotrophic gyre”); targeted secondary searches were conducted for each ocean province considered in Section 3. The search covered literature to April 2026, with subsequent targeted updates up until July 2026 to capture developments in the IMO Net-Zero Framework and related regulatory instruments. Peer-reviewed literature was prioritised for biogeochemical, ecological, and lifecycle-assessment claims; institutional, regulatory, and technical sources were retained where they constitute the primary or only available evidence, notably for regulatory instruments, engine emissions data, and production inventories, reflecting the early stage of commercial AMF deployment. Where published estimates diverge, we report ranges rather than central values and identify the assumptions driving divergence; where a more recent or better-constrained estimate is preferred over an earlier one, the basis for that preference is stated in the text. Values derived by the authors rather than taken directly from source publications—including application of IPCC default emission factors to reported leakage rates and conversion to CO₂-equivalent (CO₂e) units—are identified in the accompanying table footnotes.

1.2 Two exposure pathways

It is also notable that Nr released across the AMF value chain reaches marine ecosystems through two pathways that differ in concentration, spatial extent, duration and ecological consequence, and which are treated separately throughout this review.

First, chronic diffuse deposition refers to comparatively low deposition fluxes which are continuous, spatially extensive, and cumulative with existing anthropogenic Nr inputs. This exposure pathway is that addressed in Section 3.

Conversely, acute localised release refers to high concentration delivery of Nr directly to the water column at a point over minutes to hours, which may result, for instance, from accidental spills (e.g., during emergency response). Such acute, localized releases and their environmental dynamics have been reviewed elsewhere (e.g., Yeboah et al., 2026) and are not the focus of this review, with the exception of effluent discharge, which is reviewed as a governance gap in Sections 5.3 and 6.3.

Finally, the above pathways are not additive on a common basis because they accumulate in the environment in very different ways and result in distinct effects. Therefore, aggregating them into a single lifecycle Nr mass figure, as emissions inventories necessarily do, obscures this distinction. Where this review reports lifecycle Nr totals (Sections 2 and 4), those figures serve to characterise the scale of nitrogen mobilised by AMF deployment and should not be read as a proxy for ecological impact, which depends on the pathway, location, and form of release.

2 Lifecycle Nr emissions from AMF

AMF introduces a wider range of Nr emissions than conventional marine fuels, for which NO_x emissions dominate total Nr release (Valera-Medina et al., 2018). In contrast, Nr emissions from AMF include NO_x, N₂O and NH₃, which can be direct (e.g., NO_x and N₂O from combustion, NH₃ slip, and NH₃ from spills or accidents) and indirect (e.g., indirect N₂O), the latter arising through subsequent nitrogen-cycle transformations with long-term climate and ecosystem impacts (Galloway et al., 2003; Bertagni et al., 2023; Esquivel-Elizondo et al., 2025). The following subsections synthesise current understanding of Nr emissions across the main AMF lifecycle stages: production, storage and bunkering, and shipboard combustion.

2.1 Production-stage Nr emissions

Despite mature infrastructure, established regulation, and increasing satellite-based emissions monitoring, production facilities remain a significant source of Nr emissions. Nr emissions vary markedly by production route, with coal-based (“black”) pathways at the high end and renewable electrolysis-based (“green” or “e-”) pathways at the low end (Box 1). As production capacity expands to meet projected shipping and energy demand (approximately 230 Mt yr^-1 for shipping fuel and a further 300 Mt yr^-1 across existing agricultural and emerging power-generation uses by 2050), total emissions will scale accordingly (IEA, 2023).

2.1.1 NO_x emissions across production routes

NO_x emissions vary substantially across NH₃ production pathways. Black NH₃ production exhibits the highest NO_x emissions (2.0–4.5 kg N tonne^-1 NH₃), reflecting elevated combustion temperatures, greater energy intensity, and fuel nitrogen (fuel-N) in coal (European Commission, 2007; Bicer and Dincer, 2018). Grey NH₃ emits approximately 0.9–2.8 kg N tonne^-1 NH₃, while blue NH₃ reduces emissions to approximately 0.2–1.5 kg N tonne^-1 NH₃, likely reflecting plant configuration and inventory assumptions (IEA, 2021; Schmidt et al., 2025).

Renewable NH₃ production can reduce NO_x emissions by 90–98%, typically to less than 0.1 kg N tonne^-1 NH₃, relative to fossil-based pathways when entirely powered by renewable electricity, avoiding almost all combustion-related NO_x other than minor infrastructure-, manufacturing-, and construction-related emissions (e.g., wind turbine manufacturing, solar panel production, and hydroelectric facility construction) (European Commission, 2007; IEA, 2021; Schmidt et al., 2025). For renewable-based NH₃ made with electricity from the grid (Box 1), however, NO_x emissions remain sensitive to grid composition and biomass use, potentially approaching grey NH₃ levels if unmanaged (European Commission, 2007; IEA, 2021; Schmidt et al., 2025).

2.1.2 N₂O emissions across production routes: direct and indirect pathways

Although NO_x dominates the Nr profile of NH₃ production, N₂O emissions warrant separate consideration given its distinct climate and ozone-depleting impacts (Ravishankara et al., 2009). Production-stage N₂O arises from direct pathways associated with combustion and energy supply and indirect pathways associated with atmospheric nitrogen deposition.

The core Haber-Bosch reaction generates negligible direct N₂O because it is a reductive rather than an oxidative pathway (IEA, 2021). Direct production-stage N₂O instead arises primarily from auxiliary combustion units (steam methane reformer [SMR] furnaces, process heaters, auxiliary boilers) via fuel-N oxidation and incomplete combustion. Emissions vary by roughly an order of magnitude across routes (Table 2). Black NH₃ sits at the upper end, reflecting higher combustion temperatures, coal nitrogen content, and gasification chemistry in which N₂O can form as an intermediate species (Bicer and Dincer, 2018). Grey and blue NH₃ are comparable because carbon capture targets CO₂ and leaves combustion-related N₂O largely unchanged. Across all routes, production-stage processes account for 0–27% of lifecycle direct N₂O, with the remainder arising from downstream combustion (Esquivel-Elizondo et al., 2025).

Upstream fuel supply chains contribute to smaller but non-negligible N₂O emissions (Table 2). For grey and blue NH₃, emissions arise from compressor stations, gas processing, and pipeline combustion, though fugitive methane (CH₄) dominates the non-CO₂ upstream climate impact (IEA, 2021; Lee et al., 2022). Coal supply chains add comparable or higher contributions depending on mine ventilation and transport distances (Bicer and Dincer, 2018). For renewable-based NH₃, upstream N₂O is associated with renewable electricity infrastructure manufacturing and remains low for dedicated wind or solar systems. Upstream N₂O may increase substantially for grid-connected facilities that draw fossil-derived electricity (Schmidt et al., 2025).

Furthermore, a fraction of facility NO_x emissions are deposited within 50–500 km of the source and undergoes microbial conversion to N₂O in soils and surface waters, while chemically transformed NO_x can travel farther before depositing (Dentener et al., 2006). Applying the IPCC default emission factor (1% of deposited N; range 0.2–1.8%) to the NO_x ranges in Section 2.1 yields the indirect N₂O contributions shown in Table 2 (Hergoualc’h et al., 2019).

A second indirect pathway operates through fugitive NH₃ from production facilities, which are systematically underestimated in bottom-up inventories, as confirmed by satellite observations (Van Damme et al., 2018). As a result, Esquivel-Elizondo et al. (2025) adopt facility NH₃ leakage rates of 0.01–0.5% of production—higher than the 0.01–0.02% implied by inventories. Applying the IPCC 1% deposition default emission factor to this range yields the indirect N₂O contribution shown in Table 2, which, at the upper bound, becomes a major contributor to total production-stage N₂O. At coastal facilities, conversion efficiency is more variable and may be substantially higher in OMZ-adjacent waters (Section 3.3).

Summed across direct combustion sources, upstream supply chains, and indirect formation from on-site NO_x and fugitive NH₃ deposition, production-stage N₂O totals roughly 0.09–0.31 kg N₂O tonne^-1 NH₃ for black NH₃, 0.03–0.18 kg N₂O tonne^-1 NH₃ for grey and blue NH₃ pathways and 0.001-0.07 kg N₂O tonne^-1 NH₃ for renewable-based NH₃ (Table 2) This corresponds to climate penalties of approximately 1.3–4.5 g CO₂e MJ^-1 for black NH₃, 0.4–2.6 g CO₂e MJ^-1 for grey and blue NH₃, and 0.01-1.0 g CO₂e MJ^-1 for renewable-based NH₃. The renewable-based NH₃ upper bound assumes that future facilities exhibit fugitive NH₃ leakage comparable to existing grey and blue plants—a precautionary assumption that may overstate actual performance if new NH₃ facilities achieve tighter containment, but one that is not yet verifiable from operational data at commercial scale.

2.2 Storage, transport, and bunkering NH₃ emissions

Storage, transport, and bunkering operations generate fugitive NH₃ emissions through tank breathing (diurnal temperature cycles causing vapor expansion or contraction), pressure relief events (safety valve releases during overpressure), loading and unloading operations (displacement of vapor space), pipeline and transfer system leaks (flanges, valves, pumps), and emergency venting (equipment failures, procedural violations) (European Environment Agency, 2023; MMMCZCS, 2023). Modern NH₃ storage facilities equipped with vapor recovery systems, pressure-temperature control, and leak detection achieve emission rates of 0.01–0.05% of throughput under normal operation, equivalent to 0.1–0.5 kg NH₃ tonne^-1 across typical multi-transfer supply chains (i.e., production, storage, pipeline/ship transport, port storage, and ship bunkering) (European Environment Agency, 2023). In contrast, older or inadequately maintained facilities lacking vapor recovery systems can exhibit substantially higher emission rates, with episodic releases during equipment failures, procedural errors, or safety system activations generating short-duration high-concentration events with local air quality and safety implications (European Environment Agency, 2023).

Boil-off gas (BOG) arising from heat ingress into cryogenic storage tanks or transfer systems is the dominant source of operational NH₃ emissions during bunkering and storage, with estimated daily rates of 0.03–0.07% depending on tank size, insulation quality, and surface-area-to-volume ratio compared to ~0.2% for liquefied natural gas (LNG) (Al-Breiki and Bicer, 2020; Esquivel-Elizondo et al., 2025). Relatedly, modelled bunkering emissions (0.07–0.13 Tg NH₃-N yr^-1 by 2050) (Esquivel-Elizondo et al., 2026a) are substantially lower than earlier LNG-analog estimates (1.18–2.32 Mt NH₃ yr^-1) (Chen et al., 2024), reflecting different assumed engineered-system performance for AMF due to its physicochemical properties (e.g., liquefaction temperature of -33 °C vs. -161.5 °C for LNG) and higher toxicity (Esquivel-Elizondo et al., 2026b). Achieving near-zero bunkering emissions, however, requires sustained sensor readings consistently below 2 ppm within 2 m of transfer systems—well below occupational exposure limits (25 ppm short-term, 10 ppm long-term)—establishing a measurement-based threshold for safe bunkering design (Esquivel-Elizondo et al., 2026a).

Beyond these routine operational pathways, accidental releases of NH₃ during production, storage, bunkering, or in-transit operation constitute an additional episodic Nr input that falls outside normal lifecycle accounting; historical incident analysis indicates that human error and equipment failure are the dominant causes (Yeboah et al., 2026).

2.3 Combustion-stage Nr emissions: NO_x and N₂O formation and NH₃ slip

Near-term commercial deployment of AMF relies on dual-fuel internal combustion engines (ICEs) that use hydrocarbon pilot fuel for ignition. Shipboard combustion in ICEs is the dominant lifecycle Nr source, generating NO_x through thermal fixation and fuel-N oxidation, while also producing NH₃ slip from incomplete combustion. Fuel cells offer a near-zero alternative but remain costly and insufficiently mature for large vessels (Fullonton et al., 2025).

Untreated AMF-engine NO_x emission factors span approximately 2–9 g NO_x MJ^-1 (~38–169 kg NO_x tonne^-1 NH₃), depending on engine architecture and operating conditions, with pure NH₃ engines towards the lower end (~7 g NO_x kWh^-1) and NH₃-H₂ dual fuel engines toward the higher end (~33 g NO_x kWh^-1) (Wong et al., 2024) This range is comparable to or exceeds untreated NO_x emissions from conventional heavy fuel oil (~4.7 g MJ^-1), because the fuel-bound nitrogen pathway in NH₃ combustion substitutes for the thermal NO_x that would otherwise be limited by NH₃’s lower flame temperature (Valera-Medina et al., 2018). SCR treated AMF engines can reduce NOx emissions below Tier III compliance which is <0.5 g NOx MJ^-1 fuel (Rouwenhorst, 2025). NH₃ also presents distinct combustion challenges relative to fossil fuels: slow laminar flame speed (7 cm s^-1 versus 37 cm s^-1 for LNG [CH₄]), high auto-ignition temperature (651 °C versus 580 °C for diesel), and narrower flammability limits (16–25% by volume) (Valera-Medina et al., 2018; Kobayashi et al., 2019).

N₂O emissions from AMF ICEs remain incompletely constrained pending operational data, as AMF ICEs are still under development at the time of writing (Schmidt et al., 2025). Initial measurements from four-stroke engines indicate N₂O emissions ranging from 0.1 to 3.8 kg N₂O tonne^-1 NH₃ (equivalent to 0.008–0.29% of fuel-N emitted as N₂O), while numerical simulation studies report values from negligible under optimized conditions up to 4.9 kg N₂O tonne^-1 NH₃ (0.379% of fuel-N), with the highest simulated values reported for two-stroke engines (up to 4.9 kg N₂O tonne^-1 NH₃; 0.379% of fuel-N) and slightly lower maxima for four-stroke engines (up to 3.6 kg N₂O tonne^-1 NH₃; 0.278% of fuel-N) (Pacific Environment et al., 2025).

Incomplete or poorly controlled combustion can lead to NH₃ slip, arising from fuel-rich zones, flame quenching at combustion chamber walls, and catalyst slip, where injected NH₃ or engine-out NH₃ exceeds available NO_x for reaction (Valera-Medina et al., 2018). Slip rates (ranging from <0.5% to 2–5% of fuel mass) are highly sensitive to combustion conditions, with stoichiometric to slightly lean operation minimizing slip, very lean operation increasing slip through reduced flame temperature and slower kinetics, and rich operation exponentially increasing slip as oxygen becomes limiting (Kobayashi et al., 2019). Engine load variations typical of marine operations from low-power manoeuvring to high-load open-ocean transit create transient slip events as the fuel-air mixture and combustion temperature fluctuate faster than control systems can respond. This mechanism is well-documented for CH⁴ slip in LNG dual-fuel engines where turbocharger lag disrupts air-fuel ratio during load changes (Stoumpos et al., 2020; Kuittinen et al., 2024) and is applicable by analogy to AMF, given its slower combustion kinetics and narrower flammability limits (Valera-Medina et al., 2018; Kobayashi et al., 2019).

Emissions control in engines involves a persistent trade-off between minimizing NO_x formation and limiting NH₃ slip: higher temperatures improve the completeness of NH₃ oxidation but increase thermal NO_x, whereas lower temperatures reduce NO_x but increase NH₃ slip and other incomplete combustion byproducts (Wu et al., 2023; Xu et al., 2023; Cameretti et al., 2025; Yang et al., 2026). Similarly, selective catalytic reduction (SCR) with excess urea reduces NO_x at the expense of increased NH₃ slip (Valera-Medina et al., 2018; Kobayashi et al., 2019). Accordingly, Nr emissions are coupled, and measures targeting one pollutant can shift emissions toward others, necessitating integrated strategies rather than single-pollutant optimization (Sections 5.3, 6.3) (MMMCZCS, 2023).

3 Marine ecosystem impacts of Nr inputs

Shipping contributes approximately 15–20% of current atmospheric nitrogen deposition, with concentrations highest along major corridors connecting the Suez Canal to European ports and along heavily-trafficked coastal routes (Gerasopoulos et al., 2006). Spatially resolved modelling indicates that under weak emission controls, AMF could substantially increase shipping’s contribution to Nr emissions, with the largest air quality impacts simulated over East Asia, North Africa, the Mediterranean Basin, and Southeast Asia (Wong et al., 2024).

This section addresses ecosystem responses to chronic atmospheric deposition (Section 1.2); acute exposure from spills and effluent discharge follows different dose–response relationships and is not covered by the enrichment thresholds discussed here.

The marine transformation of deposited Nr distinguishes AMF from both terrestrial NH₃ systems and conventional fossil marine fuels. Ship NO_x oxidises to nitric acid and nitrate aerosols on timescales of hours to days, depositing over 100s of km downwind, while NH₃ slip converts to ammonium aerosols and deposits closer to source within 10s of km, producing localised hotspots along shipping corridors (Dentener and Crutzen, 1994; Corbett et al., 1999). Once deposited, Nr enters the marine nitrogen cycle, in which the microbial processes of assimilation, nitrification, denitrification, and anammox partition between productivity, export, and gaseous loss as N₂ or N₂O (Gruber and Galloway, 2008) (Box 2).

Atmospheric Nr deposition from AMF may substantially alter marine ecosystem structure and function across diverse ocean provinces, potentially impacting their ability to provide ecosystem services such as primary production and the provision of critical nearshore habitat, with cascading impacts on marine food webs and fisheries (Sun et al., 2024). However, pathways for ecosystem impacts will be spatially variable depending on ocean chemistry, ecosystem structure, and cumulative threat loading, necessitating assessment of the direction and magnitude of potential impacts at the ocean province scale. Figure 3 provides a conceptual overview of the receiving provinces examined in this section—oligotrophic gyres, coral reefs, OMZs, coastal upwelling regions, and marginal seas—and highlights the disproportionately high N₂O yields associated with oxygen-depleted subsurface waters (typically an order of magnitude or greater, with considerable variability).

Figure 3

3.1 Oligotrophic ocean gyres: nitrogen limitation and ecosystem sensitivity

The five subtropical and tropical ocean gyres collectively cover ~40% of the global ocean surface (150 million km²) and constitute the Earth’s largest continuous ecosystem (Karl et al., 2001; Moore et al., 2013). These systems are severely nitrogen-limited, with surface nitrate concentrations typically <10–200 nM—approximately 100-fold lower than coastal waters—and support low productivity (50–150 mg C m^-2 d^-1) and picophytoplankton-dominated communities (Prochlorococcus, Synechococcus) adapted over millions of years to chronic scarcity through high-affinity nitrate transporters and nitrogen-efficient photosynthetic machinery (Moore et al., 2013).

This adaptation makes oligotrophic gyres sensitive to atmospheric nitrogen deposition. Modelling studies indicate that in the absence of co-limitation by other nutrients including iron or phosphorus, moderate nitrogen enrichment comparable to projected shipping-related impacts (0.1–0.3 g N m^-2 yr^-1) could enhance primary productivity by 5–15% and carbon export 3–10%, while larger enrichment (>0.5 g N m^-2 yr^-1) shifts communities toward larger phytoplankton with reduced nitrogen-use efficiency and decreased net carbon sequestration (Duce et al., 2008; Krishnamurthy et al., 2009). However, the direction and magnitude of ecological responses to Nr inputs remain highly uncertain and depend on ocean chemistry and biogeography (Browning et al., 2017).

Several major shipping corridors intersect these sensitive regions, including the trans-Pacific route (current atmospheric deposition of 0.1–0.3 g N m^-2 yr^-1), and trans-Atlantic and Caribbean routes (0.2–0.4 g N m^-2 yr^-1) crossing the North Pacific and North Atlantic subtropical gyres, respectively (Duce et al., 2008; Jickells et al., 2017). Under high penetration AMF scenarios with moderate emission controls, spatial modelling indicates that shipping’s contribution to atmospheric Nr loading along the North Pacific and North Atlantic corridors could increase substantially (Wong et al., 2024). This increase would expose oligotrophic gyres to nitrogen inputs exceeding historical baselines, with unknown but potentially large impacts on primary production rates, phytoplankton community composition, and pelagic ecosystem structure and function.

3.2 Coral reef ecosystems: nitrogen enrichment and ecological degradation

Zooxanthellate corals form shallow water reefs which support ~25% of marine biodiversity while occupying <0.1% of the ocean surface, and their adaptation to oligotrophic conditions makes them highly sensitive to nutrient loading. Atmospheric nitrogen deposition affects coral reef ecosystems through three interacting mechanisms: (1) disruption of coral-zooxanthellae symbiosis under elevated nutrient conditions; (2) competitive promotion of fleshy macroalgae over corals; and (3) synergistic interaction with thermal stress, reducing resilience to bleaching events (Chen et al., 2019). Field observations demonstrate that these impacts occur at concentrations achievable through atmospheric deposition.

Coral reefs are highly sensitive to nitrogen enrichment because coral-zooxanthellae symbiosis functions optimally under nitrogen-limited conditions (nitrate <1 µM, ammonium <0.5 µM), making reefs particularly vulnerable to even modest increases in Nr (Fabricius, 2005; D’Angelo and Wiedenmann, 2014). Experimental and field studies indicate chronic nitrogen enrichment of 1–3 µM can reduce coral cover, increase macroalgal abundance (Fabricius et al., 2012), elevate disease prevalence and lower thermal bleaching thresholds (Vega Thurber et al., 2014), suggesting that even modest enrichment disrupts symbiotic and competitive relationships (Fabricius, 2005; D’Angelo and Wiedenmann, 2014).

Exposure risk is concentrated where major shipping corridors intersect with coral reef systems, including for instance the Caribbean reefs adjacent to Panama Canal-Atlantic, the Red Sea along the Suez Canal, and offshore Great Barrier Reef Regions (Figure 3), where AMF-related deposition is already contributing to total atmospheric nitrogen inputs (Fine et al., 2013; Bell et al., 2014; Chen et al., 2019). Under high-penetration AMF scenarios, deposition could roughly double, consistent with projected ship traffic within major shipping corridors (Wong et al., 2024). Assuming a 50–100 m mixed layer, this would increase annual surface-water nitrate concentrations by approximately 0.2–0.8 µM, sufficient to measurably affect coral physiology under chronic exposure (Fabricius, 2005).

The interaction between nitrogen enrichment and thermal stress represents the most consequential pathway under climate change. Because elevated nitrogen availability reduces coral bleaching thresholds by 0.5–1.0 °C, deposition levels insufficient to cause eutrophication in isolation may nonetheless significantly increase vulnerability during thermal events (Vega Thurber et al., 2014; Hughes et al., 2017). As marine heatwaves become more frequent and intense, this interaction suggests that even moderate increases in nitrogen deposition from AMF shipping could accelerate reef degradation in already-stressed systems.

3.3 Oxygen minimum zones: biogeochemical amplification and climate feedback

OMZs only occupy around 8% of ocean volume, yet they account for 30–50% of global oceanic nitrogen loss and roughly half of marine N₂O production, making them major hotspots of indirect N₂O formation under low-oxygen conditions (Codispoti, 2010; Naqvi et al., 2010; Sun et al., 2024), depending on the oxygen threshold used to define oxygen-deficient waters. High-density shipping corridors overlap with several major OMZs (e.g., eastern tropical Pacific, Arabian Sea, Bay of Bengal, and southwestern African margin; Figure 3), increasing the likelihood that AMF-derived nitrogen deposition can be efficiently converted to N₂O (Naqvi and Noronha, 1991; Oschlies et al., 2018).

The coupling between atmospheric nitrogen deposition and OMZ N₂O formation operates as a sequential but non-deterministic process. Nitric acid and nitrate aerosols formed from NO_x over regional scales, and ammonium aerosols from NH₃ more locally along shipping corridors, can elevate surface-ocean nitrate and ammonium concentrations by ~0.1–0.5 μmol L⁻¹ along major routes (Wolfram et al., 2022). Where this enrichment stimulates production, a fraction of the resulting organic matter is exported below the mixing layer; the exported fraction is itself variable and, in the productive upwelling systems overlying several major OMZs, deposited Nr represents a small increment relative to the upwelled nitrate.

The N₂O yield from remineralisation of that exported material is not a simple function of ambient oxygen. The principal microbial pathways of N₂O production and nitrogen removal are described in Box 2, with net N₂O production depending on the balance between these pathways and their different oxygen optima. Nitrification typically peaks in the upper oxycline, where oxygen is low but not absent, with per-unit yields rising as oxygen declines (Yoshida et al., 1989; Stieglmeier et al., 2014). N₂O accumulation further depends on microbial community composition and the relative expression of each step as well as on redox conditions, while the balance between anammox and denitrification further modulates net yield (Codispoti, 2010; Zhang et al., 2023; Sun et al., 2024).

Critically, N₂O is also consumed. N₂O reductase activity is favoured under near-anoxic conditions, and OMZ cores can act as net N₂O sinks even as the surrounding oxyclines act as strong sources (Codispoti, 2010; Naqvi et al., 2010; Zhang et al., 2023). Net N₂O flux from an OMZ therefore reflects the vertical and temporal juxtaposition of production and consumption zones rather than simply bulk oxygen concentration. Sharp gradients, mesoscale variability, and intermittent ventilation that repeatedly move water parcels across the oxygen range favouring incomplete denitrification can generate disproportionate N₂O relative to steady-state conditions at the same mean oxygen concentration. Anoxic microenvironments within sinking particles additionally support denitrification and N₂O production in waters that are oxygenated at the bulk scale (Wan et al., 2023). This further decouples yield from measured dissolved oxygen.

Empirical yields consequently span a wide range and are not tightly constrained: observations from coastal hypoxic systems indicate roughly 1–4% of remineralised nitrogen released as N₂O, compared with approximately 0.1–0.5% in well-oxygenated waters (Naqvi et al., 2010; Short et al., 2013), but reported values vary substantially within and between systems and with the oxygen threshold used to define the OMZ boundary.

The spatial heterogeneity of these yields has direct implications for NH₃ shipping. Suntharalingam et al. (2012) modelled marine N₂O responses to anthropogenic nitrogen deposition and found regional increases of 15–30% in coastal seas and up to 50% in the Arabian Sea, where deposition coincides with hypoxic waters (i.e., with moderate oxygen depletion). By contrast, increases in well-oxygenated open-ocean regions were typically <5%. Although the global integrated effect was modest (3–4% of total ocean N₂O production), the strong regional amplification over OMZs is the relevant signal for shipping, because corridor geography concentrates deposition over precisely these sensitive zones (Figure 3).

For illustrative purposes, the potential magnitude of this pathway can be bounded. Applying an OMZ-relevant conversion of 1–4% of deposited nitrogen to N₂O-N—broadly consistent with the observational yields reported for hypoxic and oxygen-depleted systems above (Codispoti, 2010; Naqvi et al., 2010; Short et al., 2013)—gives 10–40 kg N₂O-N, equivalent to approximately 16–63 kg N₂O, tonne^-1 of Nr deposited. At a 100-year global warming potential of 273 (Forster et al., 2021), this corresponds to approximately 4.3–17.2 tonne CO₂e tonne^-1 of Nr deposited.

Expressed per unit of delivered energy, lifecycle Nr emissions from AMF span 0.2–1.9 g N MJ^-1 across low- to high-emission scenarios, of which 94–98% is deposited NH₃-N and NOx-N; the corresponding figure for VLSFO is 0.68 g N MJ^-1, almost entirely as combustion NOx-N (Esquivel-Elizondo et al., 2026a). Applying the same conversion range to both fuels gives indirect N₂O forcing of approximately 1–31 g CO₂e MJ^-1 for AMF and 2.9–11.6 g CO₂e MJ^-1 for VLSFO. The implications of these indirect climate impacts, which fall outside conventional GHG accounting, are discussed in Sections 4.3 and 6.4.

These figures are indicative rather than predictive and should be read as an upper envelope. They assume that deposited Nr reaches waters where N₂O production exceeds consumption, whereas only a fraction of emitted Nr deposits over oxygen-depleted regions and only a fraction of that reaches the depths and residence times at which incomplete denitrification dominates. They also apply a uniform conversion across systems in which yields vary by an order of magnitude and depend on microbial community structure, substrate supply, and oxygen variability as well as on mean oxygen concentration. Because the same factor is applied to both fuels, the relative comparison is less sensitive to that uncertainty than the absolute values. Constraining the deposited fraction and its province-specific conversion at the ocean-province scale remains a first-order research priority (Section 4.5).

3.4 Coastal upwelling regions and marginal seas: cumulative stress and eutrophication amplification

Coastal upwelling regions (e.g., the California Current, Humboldt Current, Canary Current, and Benguela Current) and enclosed or semi-enclosed marginal seas (e.g. the Baltic Sea, Mediterranean Sea, Black Sea, North Sea, East China Sea, Yellow Sea, and South China Sea) present distinct vulnerability patterns from oligotrophic gyres and coral reefs. These systems already are characterized by naturally high nitrogen availability from upwelling or riverine inputs, existing eutrophication pressures from terrestrial nutrient loading, dense human populations in adjacent coastal zones, and intensive fishing that has altered food web structure, and ecosystem resilience (Diaz and Rosenberg, 2008; Rabalais et al., 2010).

In these systems, AMF-related Nr deposition would primarily compound existing eutrophication stress rather than introduce a novel stressor (Rabalais et al., 2010; Short et al., 2013). This is particularly important in enclosed or semi-enclosed seas, including the Baltic, Mediterranean, and East Asian marginal seas, where the combined effects of terrestrial runoff, wastewater discharge, and atmospheric deposition from shipping exceed the seas’ assimilative capacity and risk impacting critical nearshore ecosystems such as seagrass beds and their services, including coastal carbon storage and fishery production (Schmidt et al., 2012). Excess nitrogen can promote fast-growing ephemeral macroalgae and epiphytes that shade seagrass leaves, reducing meadow condition and associated ecosystem services (Burkholder et al., 2007; Schmidt et al., 2012), while increased organic matter export to sediments may further affect benthic-pelagic coupling and alter carbon and nutrient transformations in seagrass sediments and associated coastal habitats (Jiang et al., 2022).

The Baltic Sea exemplifies cumulative nitrogen stress in an enclosed marginal sea: total nitrogen inputs from atmospheric deposition (25–30% of loading), riverine discharge (60–65%), and direct point sources (5–10%) collectively deliver approximately 600–800 thousand tonnes nitrogen annually to a sea with a surface area of 420,000 km² and limited water exchange with the North Atlantic (Andersen et al., 2017; HELCOM, 2018). This loading, equivalent to 1.4–1.9 g N m^-2 yr^-1 as direct deposition plus additional inputs from river discharge, substantially exceeds the assimilative capacity of the Baltic Sea, resulting in chronic eutrophication, summer algal blooms (including toxic cyanobacteria), oxygen depletion in deep basins affecting 60,000–100,000 km² of seabed, and impaired ecosystem services including reduced fish populations and beach fouling (HELCOM, 2018). Shipping already contributes substantially to atmospheric nitrogen inputs in the Baltic region—approximately 20–25% of total atmospheric deposition originates from international shipping transiting the Baltic and North Seas, with additional contributions from coastal shipping and port activities (Jonson et al., 2015). AMF deployment along these heavily trafficked routes could substantially increase shipping’s contribution to atmospheric Nr deposition under weak emission control scenarios (Wong et al., 2024).

The Mediterranean Sea demonstrates similar cumulative nitrogen stress patterns but under different oceanographic conditions. Strong density stratification, limited freshwater inputs, high evaporation, and warm temperatures reduce vertical nutrient supply, thereby increasing the relative importance of atmospheric deposition compared with upwelling (Markaki et al., 2003; Krom et al., 2004). Total nitrogen inputs to the Mediterranean average approximately 0.8–1.2 g N m^-2 yr^-1 from atmospheric deposition (40–45%), riverine discharge (45–50%), and other direct inputs (5–10%), with substantial spatial variation between the oligotrophic eastern basin and the more enriched western basin receiving Rhône River discharge and Strait of Gibraltar inflow (Markaki et al., 2003; Krom et al., 2004). Unlike temperate marginal seas, where excess nitrogen often causes deep-water hypoxia, Mediterranean impacts are more expressed through surface productivity. This vulnerability extends to Mediterranean coral and seagrass ecosystems, including endemic Posidonia oceanica seagrass meadows and mesophotic coral communities, which exhibit nitrogen sensitivity comparable to that of tropical reefs, meaning that atmospheric deposition may affect high-value habitats disproportionately relative to overall basin productivity (Lejeusne et al., 2010).

East Asian marginal seas—the East China Sea, Yellow Sea, and South China Sea—collectively represent one of the world’s most heavily-trafficked shipping regions (Figure 3) (Corbett and Koehler, 2003). These areas already receive atmospheric nitrogen deposition of 0.3–1.0 g N m^-2 yr^-1 from combined industrial, agricultural, and shipping emissions, which has shifted some systems from being nitrogen-limited to phosphorous-limited (Zhang et al., 2010; Kim et al., 2011). These seas support the world’s most productive marine fisheries (approximately 20% of global fish catch), but face severe ecosystem impacts, including eutrophication, harmful algal blooms, expanding hypoxia, and declining fish stocks attributed to cumulative stress from nutrient loading, overfishing, pollution, and coastal habitat destruction (Zhang et al., 2010). Current atmospheric nitrogen deposition contributes approximately 20–30% of total nitrogen inputs to these systems, with shipping accounting for 15–25% of atmospheric deposition depending on location relative to major ports (Shanghai, Singapore, Hong Kong, Busan) and shipping lanes (Harrison et al., 2008).

Enhanced nitrogen loading in marginal seas may exacerbate coastal eutrophication, increasing oxygen depletion and associated impacts on fisheries and marine ecosystems, potentially reducing fishery biomass and shrinking the areas suitable for commercial fishing (Bauer et al., 2018). These low-oxygen conditions can also promote indirect N₂O formation (up to ~1–2%, Wang et al., 2016), or even higher, in stratified interfaces between oxygenated and deoxygenated waters Tan et al. (2025), where nitrification and denitrification can occur in close proximity, including within suspended particles (Wan et al., 2023). Although shallower and more seasonal than the open-ocean OMZs discussed in Section 3.3, these systems are disproportionately active N₂O sources relative to their area and are expanding under continued eutrophication pressure (Naqvi et al., 2010).

4 Discussion

4.1 AMF as a conditionally sustainable fuel option

The reviewed literature shows that AMF is not inherently low-emissions; its performance depends on the production pathway and Nr emissions control across production, the supply chain, combustion, and downstream pathways (Bertagni et al., 2023; Esquivel-Elizondo et al., 2026a). Under stringent controls, including renewable-based production pathways, low NO_x and N₂O from combustion, and minimal NH₃ slip, AMF can substantially reduce GHGs and other pollutants relative to conventional fuels (Wong et al., 2024; Schmidt et al., 2025). Under weak controls, however, it can increase Nr emissions and shift pollution between environmental compartments, with implications for air quality, climate, and marine ecosystems (Wolfram et al., 2022; Bertagni et al., 2023).

4.2 Geographic heterogeneity of marine impacts

Marine impacts are not globally uniform. They are shaped by shipping corridors traversing provinces with widely different biogeochemical and ecological profiles, so identical emission rates can produce very different outcomes depending on where they are deposited. In oligotrophic gyres, deposited Nr is rapidly assimilated, with variable productivity and community-structure responses dominating outcomes. In coral reef systems, even modest enrichment can disrupt symbiotic and competitive relationships. In OMZs, incomplete denitrification can amplify N₂O formation, whereas in coastal upwelling and marginal seas, additional Nr is superimposed on already-stressed systems, where eutrophication and episodic hypoxia dominate the response.

No single emission factor or ecosystem metric adequately captures AMF’s marine impact. The same tonne of deposited Nr can stimulate primary production in one province, amplify N₂O formation in another, accelerate coral degradation in a third, and worsen coastal hypoxia in a fourth. The heterogeneity in marine ecosystem sensitivity, combined with the distribution of shipping corridors across multiple biogeochemical provinces, creates a regulatory challenge.

In sensitive receiving ecosystems such as coral reefs and oligotrophic gyres—and in many coastal and marginal seas—ecological vulnerability and existing anthropogenic pressures heighten the importance of effectively controlling Nr emissions from AMF. Vulnerable areas include coral reef regions along the Suez-Red Sea, Panama-Caribbean, and Great Barrier Reef corridors; the North Pacific and North Atlantic subtropical gyres, including the Sargasso Sea; coastal upwelling regions including the California, Humboldt, Canary, and Benguela Currents; and marginal seas including the Baltic, Mediterranean, Black, North, East China, Yellow, and South China Seas (Figure 3). These areas contribute disproportionately to global marine biodiversity and fisheries production, heightening the potential scale of unregulated AMF impacts on global ocean health and food production.

4.3 Indirect N₂O as a critical climate penalty

OMZ-driven N₂O formation is a potentially dominant component of AMF’s lifecycle GHG emissions when shipping corridors and AMF deposition intersect oxygen-depleted waters. Under weak controls, it could offset a substantial share of AMF’s nominal climate benefit (up to 40%) (Bertagni et al., 2023; Esquivel-Elizondo et al., 2025). Even as further research is needed to refine marine conversion factors, studies which integrate indirect N₂O as a critical component of lifecycle climate performance emphasise that the climate case for AMF is strongest under deployment pathways that keep lifecycle NO_x and NH₃ emissions low enough to limit downstream N₂O formation.

This dependence on marine conversion factors further distinguishes AMF from terrestrial agricultural NH₃. Agricultural indirect N₂O emissions primarily form in soils, with empirical emission factors reasonably well constrained by field measurements (0.1–7.3%) (Mosier et al., 2004; Hergoualc’h et al., 2019), although values of up to 21% have been reported (Denmead et al., 2010; Signor et al., 2013). By contrast, marine deposition is subject to orders-of-magnitude variability in biogeochemical response across ocean provinces. As a result, identical emission rates may lead to substantially different indirect impacts depending on corridor geography (Section 3.2). This suggests that a single conversion factor is not appropriate for marine systems: higher values (e.g., ~1–4%) are more relevant in OMZ regions (e.g., the Arabian Sea, Bay of Bengal, eastern tropical Pacific, and southwestern African margin), while lower conversion occurs in well-oxygenated marine areas.

4.4 Governance, not chemistry alone, will determine whether AMF delivers net benefit

Whether AMF follows lower-emission or higher-emission deployment pathways will depend on governance. The 2026–2028 period represents a narrow window to establish regulatory frameworks before infrastructure deployment locks in vessel and infrastructure lifetimes of 25–30 years. With AMF vessel deliveries scaling rapidly through the late 2020s, delayed regulation risks embedding high-emission systems that are costly or infeasible to retrofit. The central policy question is therefore whether early regulatory frameworks can align deployment with low-emission design and operation before those trajectories are fixed.

The challenge is therefore not only to account for lifecycle Nr emissions, but to enable effective governance across the full nitrogen cascade. Previous work has argued that Nr should be managed across the cascade rather than at individual stages in isolation (Galloway et al., 2008). Such an integrated approach would need to consider production, combustion, effluent discharge, geographic differentiation, and cross-sectoral allocation of renewable-based NH₃ between maritime, agricultural, and industrial demand (Box 1). Whether NH₃ realises its technical potential is likely to depend on how effectively these considerations are embedded in regulatory frameworks, as discussed in Section 6.

4.5 Research priorities and AMF deployment implications

Several research priorities follow directly from the gaps identified across this review. First, production-stage Nr emissions remain poorly constrained: NO_x and N₂O emission factors for grey, blue, and renewable-based NH₃ span more than an order of magnitude, and direct, harmonised facility measurements are needed before lifecycle assessments can confidently differentiate production routes on Nr terms. Second, combustion-stage emissions require direct measurement of NO_x, N₂O, and NH₃ slip from NH₃-fuelled engines under representative duty cycles—including transients, low-load operation, and aged after-treatment—alongside refined indirect-N₂O conversion factors. Third, marine ecosystem dose-response remains comparatively understudied: unlike terrestrial systems, where critical loads are well established (Bobbink et al., 2010), the empirical base for excess Nr in non-estuarine marine waters does not generalise to open-ocean release from shipping corridors (Chen et al., 2024), and geographically resolved analysis is needed in various marine ecosystems, including enclosed or semi-enclosed seas where projected deposition increases intersect with sensitive ecosystems.

Robust monitoring frameworks will be essential to empirically detect ecological responses to AMF-derived Nr deposition, distinguish anthropogenic impacts from natural variability, and identify ecological tipping points. Monitoring should integrate atmospheric deposition measurements with water-column biogeochemical indicators (e.g., dissolved inorganic nitrogen, chlorophyll-a, primary productivity, and oxygen concentrations) and ecosystem-specific biological indicators tailored to vulnerable habitats. For example, a specific priority is to better resolve benthic responses to AMF-related Nr deposition in coastal and marginal seas, where chronic Nr enrichment may alter ecosystem structure and function. Future monitoring should pair deposition and water-column eutrophication indicators with benthic indicators, including macrobenthic community composition, ecological-status indices, functional trait diversity, bioturbation potential, and the condition of blue carbon habitats, such as seagrass meadows (Dimitriou et al., 2017).

These research needs also have immediate implications for deployment. They support a precautionary approach in which scale-up is accompanied by monitoring, transparent reporting, and the capacity to update assumptions as evidence improves. At the Earth-system scale, they also align with emerging evidence that anthropogenic Nr exerts significant net climate forcing through coupled atmospheric and biogeochemical pathways (Gong et al., 2024; Tian et al., 2024), reinforcing the need to evaluate AMF in terms of lifecycle nitrogen dynamics as well as GHG performance.

The history of NH₃ as a fertiliser provides a clear warning. Although NH₃ is already produced, traded, and shipped globally, policy attention has historically focused more on production-stage emissions than on downstream Nr losses, even though those losses have driven much of the planetary-scale perturbation of the nitrogen cycle (Figure 2). A similar pattern could emerge as NH₃ is reintroduced as an energy carrier (Erisman, 2026). The “pervasive inefficiencies” identified in agricultural Nr use a generation ago (Galloway et al., 2008) have an analogue in the projected AMF value chain, where leakage and combustion losses vary by more than an order of magnitude depending on technology and operational practice (Esquivel-Elizondo et al., 2026a).

5 Mitigation strategies for Nr emissions across the AMF lifecycle

The most effective means to mitigate Nr emissions from shipping is, where technically feasible, adopting zero-emission propulsion such as wind or battery-electric operation (Kirkels et al., 2026). However, where AMF is used, minimising its marine ecosystem and wider impacts depends on effectively mitigating lifecycle Nr emissions. Various strategies for doing so are reviewed in this section. Mitigation options are considered through the three stages where intervention is both technically feasible and consequential for overall environmental performance: (i) NH₃ production; (ii) storage, transport, and bunkering; and (iii) shipboard combustion and use. The governance frameworks which could mandate or incentivize the uptake of these measures are discussed in Section 6.

5.1 Production-stage mitigation

Production-stage mitigation of Nr emissions can be achieved in part by transitioning from fossil-based black and grey NH₃ to NH₃ produced via renewable-powered electrolysis, thereby eliminating the N₂O and NO_x emissions associated with fossil fuel use in NH₃ production (Section 2.1) (IEA, 2021). A variety of measures can also help mitigate direct sources of atmospheric NH₃ arising from purge gas venting, equipment leaks, and intermittent startup or maintenance emissions (European Commission, 2007; European Environment Agency, 2023). Key measures include closed-loop purge gas recovery (routing vent streams back to feedstock preparation or hydrogen recovery), improved sealing at converter flanges and high-pressure separators, vapour-recovery systems on intermediate storage (European Commission, 2007; IEA, 2021), and continuous fence-line monitoring linked to facility-level reporting (Van Damme et al., 2018; Bertagni et al., 2023).

5.2 Storage, transport, and bunkering management

In modelled AMF value chains, these stages account for up to 78% of total NH₃ losses, depending on voyage length and BOG management (Esquivel-Elizondo et al., 2025). Achieving near-zero supply-chain emissions, therefore, depends on coordinated control across tank design, BOG management, transfer operations, and continuous leak detection.

Tank design and pipeline infrastructure. Modern cold NH₃ storage with vacuum-jacketed or perlite-insulated double-walled tanks can reduce heat ingress and BOG substantially relative to single-walled designs. For pipeline transfer infrastructure, leaks can be mitigated through high-integrity flanges, double-block-and-bleed valves, and routine pressure testing (Fertilizers Europe, 2013).

Boil-off gas recapture. Full BOG recapture and reuse—through reliquefaction with return to storage or combustion as engine fuel—eliminates this source entirely. Modelling indicates that switching from no BOG recapture to full recapture reduces total value-chain NH₃ emissions by 30–50% under medium-emissions scenarios, with corresponding reductions in indirect N₂O climate impact of 1–18 g CO₂e MJ^-1 (Esquivel-Elizondo et al., 2025). Vapour recovery systems linking storage tanks to liquefaction units enable continuous BOG capture rather than periodic venting (Bouabidi et al., 2021). The relative impact and importance of BOG recapture for reducing NH₃ emissions increases with transit duration, becoming particularly important for voyages exceeding four weeks (Lee et al., 2022; Ji et al., 2025).

Closed-loop transfer and bunkering containment. Ship-to-ship and shore-to-ship bunkering operations could utilise closed-loop transfer systems with vapour return lines, dry-disconnect couplings, and inerted manifolds to prevent emissions during line filling, transfer, and disconnection. Containment systems, including drip trays, minimise NH₃ contact with surface waters, while water curtains around transfer manifolds limit atmospheric dispersion of any release (Coates et al., 2022; MMMCZCS, 2023). Pre-cooling of bunker lines reduces flash vaporisation losses during initial transfer, and shorter bunkering durations are preferred where flow rates allow to minimise cumulative BOG generation (Esquivel-Elizondo et al., 2025).

Operational protocols. Crew training addressing NH₃’s distinct toxicity profile, procedural standards for transfer operations, and emergency response protocols are essential operational complements to the technical measures (MMMCZCS and Lloyd’s Register Maritime Decarbonization Hub, 2025).

5.3 Combustion-stage mitigation and downstream management

Combustion-stage mitigation is an important lever for reducing the lifecycle impacts of AMF, as shipboard emissions display the greatest uncertainty and could account for the majority of lifecycle Nr emissions (Section 2.3). Because efforts to reduce NO_x emissions can increase NH₃ slip, and vice versa, mitigation strategies must consider overall Nr performance rather than individual pollutants in isolation. Integrated AMF-engine emission-control systems therefore rely on initial combustion optimization to minimise engine-out Nr emissions, coupled with effective downstream emissions controls such as SCR—to achieve high NO_x conversion efficiencies—and downstream NH₃-slip and N₂O catalysts (ASC)—to oxidise residual NH₃—to enable low final exhaust concentrations of all Nr species (Voniati et al., 2023; Graziano et al., 2025; Makino et al., 2025).

Downstream emission controls—including SCR, ASC, and exhaust scrubbing where deployed—can reduce gaseous emissions but shift Nr into ammonium- and nitrate-containing effluent unless these streams are captured, treated, and withheld from discharge to port waters or the marine environment. The same applies to NH₃-contaminated deck washdown and fire-fighting water (Sofiev et al., 2018; Friends of the Earth International et al., 2025; Esquivel-Elizondo et al., 2026a).

6 Policy frameworks and governance for sustainable AMF deployment

The mitigation measures outlined above can, if implemented effectively and at scale, limit the release of Nr throughout the AMF value chain to levels significantly below current emissions from conventional shipping fuels (Esquivel-Elizondo et al., 2026a). Policy and regulatory instruments—collectively, governance frameworks—may be required to incentivize or mandate the uptake of these measures.

Relevant existing and planned governance frameworks seek to address a range of issues and span multiple jurisdictions at the global, regional, and local (national and sub-national) levels. For instance, at the global level, the IMO is responsible for developing the regulatory framework for shipping. Relevant existing IMO instruments include the International Convention for the Prevention of Pollution from Ships (IMO, 2008), the International Code of Safety for Ships Using Gases or Other Low-flashpoint Fuels, which provides criteria for the use of low-flashpoint fuels like AMF (IGF Code; IMO, 2015), and the International Convention for the Safety of Life at Sea (SOLAS; IMO, 2020b). IMO Member States are also discussing the adoption of a Net Zero Framework, a set of measures to decarbonize the sector, which may incorporate a GHG pricing mechanism (IMO, 2025a). Meanwhile, a range of relevant governance frameworks also exist at regional and local levels. In the European Union (EU), for instance, FuelEU Maritime establishes declining GHG intensity targets for shipping, the EU Emissions Trading System (ETS) prices shipping’s GHG emissions, the Ambient Air Quality Directive includes provisions for both NO_x and NH₃, the Industrial Emissions Directive regulates large NH₃ production facilities, and while the Port Reception Facilities Directive governs delivery of ship-generated waste to port (European Commission, 2019, 2023, 2024a, 2024b). Bunkering and port-side handling, meanwhile, are also regulated at the local level through bunkering licences, port air-quality rules and harbour safety codes.

The above examples are selective and incomplete but highlight the breadth of governance frameworks relating to AMF and shipping’s Nr emissions. Shipping’s current governance framework was designed for an industry powered by fossil fuels and will need to be adapted for new fuels like AMF. In the following, we outline key existing governance frameworks and gaps across the AMF value chain. We briefly discuss AMF production before focusing on AMF bunkering and use onboard vessels as key determinants of Nr impacts on marine ecosystems. We conclude by examining measures that cut across the AMF value chain.

6.1 Governance of ammonia production

Production-stage Nr emissions have potential to occur far from marine environments but could still have marine ecosystem impacts due to environmental transport, coupled with the fact that these emissions account for a substantial share of NH₃’s lifecycle Nr footprint (Section 2.1).

Current governance frameworks for NH₃ production are primarily regional or national in nature. For instance, in the EU, the Industrial Emissions Directive (European Commission, 2024a) regulates large NH₃ facilities through Best Available Technologies requirements and permit-based emissions limits, while in the US, NH₃ point source emissions must be reported to the National Emissions Inventory (U.S. EPA, 2024) under the Clean Air Act and, where reporting thresholds are exceeded, to the Toxic Release Inventory under section 313 of the Emergency Planning and Community Right-To-Know Act (U.S. EPA, 2026), with the resulting data supporting state and local air quality management.

Some schemes also seek to certify the sustainability and emissions credentials of fuel production, including AMF (e.g., Ammonia Europe, n.d; ISCC, 2024; CertifHY, 2025). Certification, however, focuses almost exclusively on the GHG intensity of AMF production and/or the origin of any renewable electricity used.

This lack of consideration of lifecycle Nr performance is an issue common to several frameworks which cut across the AMF value chain, as discussed in Section 6.4. Under these frameworks, substantially different Nr emission outcomes from NH₃ production facilities are largely indistinguishable, providing little incentive to adopt mitigation measures. Integrating these Nr assessments, coupled with facility-level measurement and disclosure of production emissions of NO_x, NH₃, and N₂O on a per-tonne-NH₃ basis, could help expose plant-level variation in Nr emissions and guide mitigating action.

6.2 Governance of storage, transport and bunkering

Emissions from AMF storage, transport and particularly bunkering are likely to occur more proximally to marine environments. Of the relevant processes, bunkering and port-side handling are regulated primarily at the national and port-authority level, and existing frameworks may not adequately address NH₃’s risk profile (Green Voyage 2050, 2023; Jang et al., 2023; Esquivel-Elizondo et al., 2026b).

For instance, national port regulations for volatile organic compound emissions and ship-to-shore transfer are governed by MARPOL Annex VI for hydrocarbon fuels but have no equivalent provisions for NH₃ fugitive losses or NH₃-contaminated effluent (IMO, 2008(Annex VI, Regulation 15)). Meanwhile, voluntary IMO and industry guidance provides technical advice and templates to guide bunkering and handling, but lacks effective enforcement and verification mechanisms (e.g., ABS, 2024; ClassNK, 2025; IMO, 2025b).

More generally, a fragmented regulatory landscape consisting of variable local, national, or regional measures creates structural socio-economic risks, for instance, by encouraging vessels to bunker in jurisdictions with weak port-emission oversight to evade more stringent controls applied elsewhere (as studied in the context of climate measures, e.g.: Kotzampasakis and Wooley (2025)). This highlights the importance of developing bunkering guidelines at an international (i.e., IMO) level, and codifying these in port state regulations, as well as encouraging cooperation between different companies, ports and states (IMO, 2019). Meanwhile, incorporating storage, transport and bunkering Nr emissions into cross-cutting assessments of marine fuel performance, such as certification schemes or the lifecycle assessments (LCAs) underpinning frameworks like the IMO’s Net Zero Framework, could further encourage improved Nr emissions mitigation (see Section 6.4).

6.3 Governance of AMF use onboard ships

The measures for mitigating Nr emissions from AMF use onboard ships, outlined in Section 5.3, may be mandated or incentivized through a variety of approaches.

Currently, the primary governance levers related to emissions from ship engines are Emission Control Areas (ECAs), operationalized by engine standards (IMO, 2008 (Annex VI)). ECAs were introduced to limit ship emissions of NO_x and sulfur oxides (SO_x) in specific regions to reduce the social and environmental impacts of these air pollutants. However, ECAs currently only address the Nr species emitted by AMF engines only partially: for instance, they do not currently regulate NH₃, despite its status as a PM_2.5 precursor and bioavailable Nr source (Wong et al., 2024).

Closing these regulatory gaps through an integrated, NH₃-specific engine emission framework could be an important lever to limit the impacts of AMF’s Nr emissions (Wong et al., 2024; Esquivel-Elizondo et al., 2026a). For instance, binding limits on ship NH₃ emissions could be introduced to MARPOL Annex VI (IMO, 2008), potentially at the ≤0.5% of fuel consumption (≈≤100 ppm exhaust) level identified as both achievable (Valera-Medina et al., 2018; Graziano et al., 2025) and sufficient to effectively limit Nr emissions and their impacts (Wong et al., 2024; Esquivel-Elizondo et al., 2026b). NO_x limits for ships using AMF could also be set at levels reflecting best-available performance (on the order of ≤0.5 g MJ^-1, ≈≤2.0 g kWh^-1; see Esquivel-Elizondo et al., 2025), while Nr emissions from use onboard ships could also be integrated into the architecture of cross-cutting instruments like the IMO Net Zero Framework and EU ETS (see Section 6.4).

Controls on ship emissions require compliance either by certifying engine performance against emission standards prior to use aboard vessels, or by continuously monitoring and reporting actual ship emissions. The engine standard compliance route is currently used to control NO_x emissions from ships, but it has a range of documented shortcomings (Comer et al., 2024). If this compliance route were to be adopted for ships using AMF, it is crucial that it is robust and that measurements of the different Nr species are integrated rather than sequential, due to observations that efforts to control NO_x, NH₃ slip, and N₂O emissions result in trade-offs (Wu et al., 2023; Section 2.3). Alternatively, continuous emissions monitoring may be a more robust compliance mechanism. While not without challenges related to measurement procedures and data management, this option may be preferred given the variable nature of Nr emissions from AMF engines (e.g., see ranges in Esquivel-Elizondo et al., 2025; Pacific Environment et al., 2025).

Importantly, efforts to govern Nr emissions from the use of AMF onboard ships highlight the need for holistic measures that limit the pollution burden rather than shift it elsewhere. Conventionally-fueled vessels provide an important lesson in this regard, where the introduction of ship SO_x emission limits gave rise to the use of open-loop scrubbers, which reduced ship SO₂ emissions but simultaneously resulted in ships discharging acidic, sulfate-rich wastewater directly into the marine environment (Sofiev et al., 2018). Emission controls on AMF ships could similarly lead to the use of exhaust treatments (e.g., to remove NO_x and NH₃ slip from exhaust gases), resulting in nitrogen-based byproducts in the effluent (Section 5.3). It’s important to emphasize that allowing the discharge of these effluents would displace rather than limit the Nr pollution burden, underscoring the need for complementary and integrated governance measures.

Finally, safety and mitigation systems aboard NH₃-fuelled ships inevitably generate NH₃-contaminated wastewater during routine operations and emergency response (Section 5.3). These effluents are not adequately addressed by shipping’s existing regulatory framework, leading the IMO Sub-Committee on Pollution Prevention and Response to recognize this gap and begin work developing guidelines for managing NH₃ effluent (IMO, 2025c; Korean Register, 2026). Possible management options include allowing discharge only under certain circumstances and strict prohibition of discharge—the latter would best address the chronic, cumulative, and indirect Nr risk dimensions reviewed in Section 3.

Regardless of the waste stream or adopted approach, effective management of ammonia effluent may require monitoring and enforcement to identify and address non-compliance. This issue is not unique to AMF—illegal discharges of, for instance, oily wastewater have been documented from conventionally-fueled ships (e.g., in the Baltic Sea (HELCOM, 2019))—nor to this particular Nr source. Generally, monitoring, compliance and enforcement mechanisms can play an important role in determining the effectiveness of governance measures (e.g., Mohite and Mathew, 2025).

6.4 Governance across value chains and jurisdictions

As referred to throughout Sections 6.1–6.3, the review of governance frameworks here shows that value-chain instruments governing marine fuels—including production certification, LCA assessments, and GHG emissions pricing—were built around carbon and treat Nr inconsistently or not at all. For instance, the IMO Net-Zero Framework, the EU ETS, and the certification schemes that underpin them rely on well-to-wake GHG accounting in which Nr appears only through the climate-forcing fraction of direct N₂O emissions. The (marine) ecosystem degradation, indirect climate forcing, and air-quality externalities documented in this review fall outside these accounts, risking the prospect that these schemes will reward AMF use, which, while it may reduce direct lifecycle GHG emissions compared to fossil fuels, can increase lifecycle Nr emissions (Wolfram et al., 2022; Esquivel-Elizondo et al., 2026a).

One option to address this issue could be to incorporate Nr emissions into existing cross-cutting certification and pricing frameworks. For instance, nitrogen-intensity metrics, reduction targets, and reward thresholds could be introduced to certification schemes and policy frameworks alongside GHG equivalents, and the climate impacts of indirect N₂O could be incorporated into GHG pricing (Esquivel-Elizondo et al., 2025, 2026a). These governance frameworks could serve to mandate and/or incentivize the reduction of Nr leaks and emissions across AMF value chains.

The implications of AMF deployment could also be effectively integrated into existing marine environmental governance frameworks. Nr pollution compromises outcomes in Marine Protected Areas (Carrasco Rivera and Wenger, 2026) and is a key driver behind whether European waters meet Good Environmental Status as required by the Marine Strategy Framework Directive (e.g., European Environment Agency (EEA), 2019). Additional Nr pollution from AMF could compromise these and other marine environmental management efforts.

More generally, cross-jurisdictional and cross-value-chain governance frameworks may be required to ensure the just and equitable use of AMF. As highlighted in Sections 6.1–6.3, a fragmented regulatory landscape consisting of variable local, national, or regional measures may create structural socio-economic risks, whereby NH₃ production and activities for AMF use (e.g., bunkering) could be pushed toward regions with weaker governance, in turn heightening the environmental and social risks of Nr pollution in those jurisdictions. It is crucial that the development of AMF governance accounts for these cross-sectoral, inherently global considerations.

Recent studies highlighted above show how effectively addressing governance gaps and omissions could introduce global controls on Nr emissions from AMF, with corresponding public health, environmental, and climate benefits. It is also clear that it is not just the existence of frameworks that matters, but also their stringency and effectiveness. With robust frameworks in place, the AMF transition can be guided toward lower-emission pathways that are more consistent with global and regional sustainability objectives.

References

• 1

  ABS (2024). Ammonia bunkering: Technical and operational advisory. Available online at: https://ww2.eagle.org/en/publication-flip/ammonia-bunkering-advisory.html (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 2

  Al-BreikiM.BicerY. (2020). Technical assessment of liquefied natural gas, ammonia and methanol for overseas energy transport based on energy and exergy analyses. Int. J. Hydrogen Energy45, 34927–34937. doi: 10.1016/j.ijhydene.2020.04.181

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  Ammonia Europe (n.d). Certification. Available online at: https://www.ammoniaeurope.com/certification/ (Accessed May 27, 2026).

  • Google Scholar
  • View reference in article

• 4

  AndersenJ. H.CarstensenJ.ConleyD. J.DromphK.Fleming-LehtinenV.GustafssonB. G.et al. (2017). Long-term temporal and spatial trends in eutrophication status of the Baltic Sea. Biol. Rev.92, 135–149. doi: 10.1111/brv.12221

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  BauerB.MeierH. E. M.CasiniM.HoffA.MargońskiP.OrioA.et al. (2018). Reducing eutrophication increases spatial extent of communities supporting commercial fisheries: A model case study. ICES J. Mar. Sci.75, 1306–1317. doi: 10.1093/icesjms/fsy003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BellP. R. F.ElmetriI.LapointeB. E. (2014). Evidence of large-scale chronic eutrophication in the Great Barrier Reef: Quantification of chlorophyll a thresholds for sustaining coral reef communities. AMBIO43, 361–376. doi: 10.1007/s13280-013-0443-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BertagniM. B.SocolowR. H.MartirezJ. M. P.CarterE. A.GreigC.JuY.et al. (2023). Minimizing the impacts of the ammonia economy on the nitrogen cycle and climate. Proc. Natl. Acad. Sci.120, e2311728120. doi: 10.1073/pnas.2311728120

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  BicerY.DincerI. (2018). Life cycle assessment of ammonia utilization in city transportation and power generation. J. Clean. Prod.170, 1594–1601. doi: 10.1016/j.jclepro.2017.09.243

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  BobbinkR.HicksK.GallowayJ.SprangerT.AlkemadeR.AshmoreM.et al. (2010). Global assessment of nitrogen deposition effects on terrestrial plant diversity: A synthesis. Ecol. Appl.20, 30–59. doi: 10.1890/08-1140.1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  BouabidiZ.AlmomaniF.Al-muslehE. I.KatebahM. A.HusseinM. M.ShazedA. R.et al. (2021). Study on boil-off gas (BOG) minimization and recovery strategies from actual baseload LNG export terminal: Towards sustainable LNG chains. Energies (Basel).14, 3478. doi: 10.3390/en14123478

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  BrowningT. J.AchterbergE. P.RappI.EngelA.BertrandE. M.TagliabueA.et al. (2017). Nutrient co-limitation at the boundary of an oceanic gyre. Nature551, 242–256. doi: 10.1038/nature24063

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  BurkholderJ. M.TomaskoD. A.TouchetteB. W. (2007). Seagrasses and eutrophication. J. Exp. Mar. Biol. Ecol.350, 46–72. doi: 10.1016/j.jembe.2007.06.024

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  CamerettiM. C.De RobbioR.PalombaM.Zucareli de SouzaT. (2025). Strategies to improve ammonia combustion in a dual fuel marine engine by using CFD. Fuel381, 133440. doi: 10.1016/j.fuel.2024.133440

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  Carrasco RiveraD. E.WengerA. S. (2026). Wastewater pollution undermines coastal marine protection: Implications for 30x30 and effective conservation. Ocean Coast. Manage.276, 108150. doi: 10.1016/j.ocecoaman.2026.108150

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  CertifHY (2025). CertifHy EU RFNBO scheme documents. Available online at: https://www.certifhy.eu/eu-rfnbo-scheme-documents/ (Accessed May 27, 2026).

  • Google Scholar
  • View reference in article

• 16

  ChenM.JiangS.HanA.YangM.TkalichP.LiuM. (2024). Bunkering for change: Knowledge preparedness on the environmental aspect of ammonia as a marine fuel. Sci. Total Environ.907, 167677. doi: 10.1016/j.scitotenv.2023.167677

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  ChenX.YuK.HuangX.WangY.LiaoZ.ZhangR.et al. (2019). Atmospheric nitrogen deposition increases the possibility of macroalgal dominance on remote coral reefs. J. Geophys. Res. Biogeosci.124, 1355–1369. doi: 10.1029/2019JG005074

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  ClassNK (2025). Guidelines for safety operation for ammonia-fuelled vessels [ver1.0. Available online at: https://download.classnk.or.jp/guidelines/gl_ammonia_operation_e.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 19

  CoatesS.DawsonL.WareJ.VestL. (2022). Ammonia at sea: Studying the potential impact of ammonia as a shipping fuel on marine ecosystems. Available online at: https://www.edfeurope.org/sites/default/files/EDF-Europe-Ammonia-at-sea-FullReport.pdf (Accessed May 2, 2026).

  • Google Scholar
  • View reference in article

• 20

  CodispotiL. A. (2010). Interesting times for marine N ₂ O. Sci. (1979).327, 1339–1340. doi: 10.1126/science.1184945

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  ComerB.ÜnalanS.MaoX. (2024). Updating marine engine emission standards using real-world data: A potential update to IMO’s NOx technical code. Available online at: https://theicct.org/publication/updating-marine-engine-emission-standards-using-real-world-data-nov24/ (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 22

  CorbettJ. J.FischbeckP. S.PandisS. N. (1999). Global nitrogen and sulfur inventories for oceangoing ships. J. Geophysical Research: Atmospheres104, 3457–3470. doi: 10.1029/1998JD100040

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  CorbettJ. J.KoehlerH. W. (2003). Updated emissions from ocean shipping. J. Geophysical Research: Atmospheres108 (D20), 4650. doi: 10.1029/2003JD003751

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  D’AngeloC.WiedenmannJ. (2014). Impacts of nutrient enrichment on coral reefs: New perspectives and implications for coastal management and reef survival. Curr. Opin. Environ. Sustain.7, 82–93. doi: 10.1016/j.cosust.2013.11.029

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  DenmeadO. T.MacdonaldB. C. T.BryantG.NaylorT.WilsonS.GriffithD. W. T.et al. (2010). Emissions of methane and nitrous oxide from Australian sugarcane soils. Agric. For. Meteorol.150, 748–758. doi: 10.1016/j.agrformet.2009.06.018

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  DentenerF. J.CrutzenP. J. (1994). A three-dimensional model of the global ammonia cycle. J. Atmos. Chem.19, 331–369. doi: 10.1007/BF00694492

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  DentenerF.DrevetJ.LamarqueJ. F.BeyI.EickhoutB.FioreA. M.et al. (2006). Nitrogen and sulfur deposition on regional and global scales: A multimodel evaluation. Global Biogeochem. Cycles20, GB4003. doi: 10.1029/2005GB002672

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  DiazR. J.RosenbergR. (2008). Spreading dead zones and consequences for marine ecosystems. Sci. (1979).321, 926–929. doi: 10.1126/science.1156401

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  DimitriouP. D.PapageorgiouN.KarakassisI. (2017). Response of benthic macrofauna to eutrophication in a mesocosm experiment: Ecosystem resilience prevents hypoxic conditions. Front. Mar. Sci.4. doi: 10.3389/fmars.2017.00391

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  DuceR. A.LaRocheJ.AltieriK.ArrigoK. R.BakerA. R.CaponeD. G.et al. (2008). Impacts of atmospheric anthropogenic nitrogen on the open ocean. Sci. (1979).320, 893–897. doi: 10.1126/science.1150369

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  ErismanJ. W. (2026). Ammonia for food and fuels in a sustainable future. One Earth9, 101621. doi: 10.1016/j.oneear.2026.101621

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  Esquivel-ElizondoS.Cabbia HubatovaM.KershawJ.SartzetakisS. S.GilliamL. (2026a). Ammonia marine fuel can reduce or increase reactive nitrogen pollution depending on emissions controls. Commun. Sustainability1, 70. doi: 10.1038/s44458-026-00076-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  Esquivel-ElizondoS.OckoI. B.AlvarezR. A.TraskB.RettigI.HamburgS. P. (2026b). Ensuring the climate and environmental integrity of alternative fuels. Environ. Sci. Technol.60, 5219–5230. doi: 10.1021/acs.est.5c07609

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  Esquivel-ElizondoS.WalkowiakB.SartzetakisS. S.BumaB. (2025). Climate impact of direct and indirect N2O emissions from the ammonia marine fuel value chain. Environ. Sci. Technol. 59 (18), 9037–9048. doi: 10.1021/acs.est.4c13135

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  European Commission (2007). Reference document on best available techniques for the manufacture of large volume inorganic chemicals - ammonia, acids and fertilisers (Luxembourg). Available online at: https://bureau-industrial-transformation.jrc.ec.europa.eu/sites/default/files/2022-03/LVIC-AAF.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 36

  European Commission (2019). “ Directive (EU) 2019/883 of the European Parliament and of the Council of 17 April 2019 on port reception facilities for the delivery of waste from ships, amending Directive 2010/65/EU and repealing Directive 2000/59/EC,” in Official Journal of the European Union. Brussels: European Commision. Available online at: http://data.europa.eu/eli/dir/2019/883/2019-06-07 (Accessed May 19, 2026).

  • Google Scholar
  • View reference in article

• 37

  European Commission (2023). “ Regulation (EU) 2023/1805 of the European Parliament and of the Council of 13 September 2023 on the use of renewable and low-carbon fuels in maritime transport, and amending Directive 2009/16/EC,” in Official Journal of the European Union. Brussels: European Commision. Available online at: http://data.europa.eu/eli/reg/2023/1805/oj (Accessed May 19, 2026).

  • Google Scholar
  • View reference in article

• 38

  European Commission (2024a). “ Directive (EU) 2024/1785 of the European Parliament and of the Council of 24 April 2024 amending Directive 2010/75/EU of the European Parliament and of the Council on industrial emissions (integrated pollution prevention and control) and Council Directive 1999/31/EC on the landfill of waste,” in Official Journal of the European Union. Brussels: European Commision. Available online at: http://data.europa.eu/eli/dir/2024/1785/oj (Accessed May 19, 2026).

  • Google Scholar
  • View reference in article

• 39

  European Commission (2024b). “ Directive (EU) 2024/2881 of the European Parliament and of the Council of 23 October 2024 on ambient air quality and cleaner air for Europe,” in Official Journal of the European Union. Brussels: European Commision. Available online at: http://data.europa.eu/eli/dir/2024/2881/oj (Accessed May 19, 2026).

  • Google Scholar
  • View reference in article

• 40

  European Environment Agency (2023). EMEP/EEA air pollutant emission inventory guidebook 2023: Technical guidance to prepare national emission inventories (Luxembourg). Available online at: https://www.eea.europa.eu/en/analysis/publications/emep-eea-guidebook-2023 (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 41

  European Environment Agency (EEA) (2019). Nutrient enrichment and eutrophication in Europe’s seas: Moving towards a healthy marine environment. Available online at: https://www.eea.europa.eu/en/analysis/publications/nutrient-enrichment-and-eutrophication-in (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 42

  FabriciusK. E. (2005). Effects of terrestrial runoff on the ecology of corals and coral reefs: Review and synthesis. Mar. pollut. Bull.50, 125–146. doi: 10.1016/j.marpolbul.2004.11.028

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  FabriciusK. E.CooperT. F.HumphreyC.UthickeS.De’athG.DavidsonJ.et al. (2012). A bioindicator system for water quality on inshore coral reefs of the Great Barrier Reef. Mar. pollut. Bull.65, 320–332. doi: 10.1016/j.marpolbul.2011.09.004

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 44

  Fertilizers Europe (2013). Guidance for inspection of leak detection in liquid ammonia pipelines (Brussels). Available online at: https://www.fertilizerseurope.com/wp-content/uploads/2019/08/Guidance_for_inspection_of_and_leak_detection_in_liquid_ammonia_pipelines_FINAL_01.pdf (Accessed May 31, 2026).

  • Google Scholar

• 45

  FineM.GildorH.GeninA. (2013). A coral reef refuge in the Red Sea. Glob. Change Biol.19, 3640–3647. doi: 10.1111/gcb.12356

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  ForsterP.StorelvmoT.ArmourK.CollinsW.DufresneJ.-L.FrameD.et al. (2021). “ The Earth’s energy budget, climate feedbacks and climate sensitivity,” in Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Eds. Masson-DelmotteV.ZhaiP.PiraniA.ConnorsS. L.PéanC.BergerS. ( Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA), 923–1054. doi: 10.1017/9781009157896.009

  • CrossRef
  • Google Scholar

• 47

  FowlerD.CoyleM.SkibaU.SuttonM. A.CapeJ. N.ReisS.et al. (2013). The global nitrogen cycle in the twenty-first century. Philos. Trans. R. Soc. B. Biol. Sci.368, 20130164. doi: 10.1098/rstb.2013.0164

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  Friends of the Earth International, Pacific International, Clean Shipping Coalition, and Environmental Defense Fund (2025). PPR 13/17/4: comments on document PPR 13/17/1 [Comment submitted to the IMO PPR sub-committee. Available online at: https://cleanshipping.org/wp-content/uploads/2026/02/PPR13-17-4-Comments-on-document-PPR-13-17-1.pdf (Accessed February 18, 2026).

  • Google Scholar

• 49

  FullontonA.Lea-LangtonA. R.MaduguF.LarkinA. (2025). Green ammonia adoption in shipping: Opportunities and challenges across the fuel supply chain. Mar. Policy171, 106444. doi: 10.1016/j.marpol.2024.106444

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  GallowayJ. N.AberJ. D.ErismanJ.SeitzingerS. P.HowarthR. W.CowlingE. B.et al. (2003). The nitrogen cascade. Bioscience53, 341–356. doi: 10.1641/0006-3568(2003)053[0341:TNC]2.0.CO;2

  • CrossRef
  • Google Scholar

• 51

  GallowayJ. N.TownsendA. R.ErismanJ. W.BekundaM.CaiZ.FreneyJ. R.et al. (2008). Transformation of the nitrogen cycle: Recent trends, questions, and potential solutions. Sci. (1979).320, 889–892. doi: 10.1126/science.1136674

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  GerasopoulosE.KouvarakisG.BabasakalisP.VrekoussisM.PutaudJ.MihalopoulosN. (2006). Origin and variability of particulate matter (PM10) mass concentrations over the Eastern Mediterranean. Atmos. Environ.40, 4679–4690. doi: 10.1016/j.atmosenv.2006.04.020

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  GongC.TianH.LiaoH.PanN.PanS.ItoA.et al. (2024). Global net climate effects of anthropogenic reactive nitrogen. Nature632, 557–563. doi: 10.1038/s41586-024-07714-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  GrazianoB.TopaloglouS.PurayilA.AvvisatiI.ChristenC.BouwstraA.et al. (2025). WinGD’s ammonia engine technology for the new X-DF-A family: New engine concepts & systems. Available online at: https://wingd.com/media/lyddydbl/cimac25-x-df-a-wingds-ammonia-engine-technology.pdf (Accessed May 31, 2026).

  • Google Scholar

• 55

  Green Voyage 2050 (2023). Regulatory mapping of marine alternative fuelsInternational Maritime Organization (Imo). Available online at: https://greenvoyage2050.imo.org/alternative-marine-fuels-regulatory-mapping/ (Accessed June 9, 2026).

  • Google Scholar

• 56

  GruberN.GallowayJ. N. (2008). An earth-system perspective of the global nitrogen cycle. Nature451, 293–296. doi: 10.1038/nature06592

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57

  HarrisonP. J.YinK.LeeJ. H. W.GanJ.LiuH. (2008). Physical–biological coupling in the pearl river estuary. Cont. Shelf Res.28, 1405–1415. doi: 10.1016/j.csr.2007.02.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  HELCOM (2018). State of the Baltic Sea — Second HELCOM holistic assessment 2011–2016 (Helsinki). Available online at: https://helcom.fi/baltic-sea-trends/holistic-assessments/state-of-the-baltic-sea-2018/ (Accessed May 31, 2026).

  • Google Scholar

• 59

  HELCOM (2019). Annual report on discharges observed during aerial surveillance in the Baltic Sea 2018 (Helsinki). Available online at: https://helcom.fi/publications/helcom-annual-report-on-discharges-observed-during-aerial-surveillance-in-the-baltic-sea-2018/ (Accessed July 21, 2026).

  • Google Scholar

• 60

  Hergoualc’hK.AkiyamaH.BernouxM.ChirindaN.del PradoA.KasimirÅ.et al. (2019). “ N2O emissions from managed soils, and CO2 emissions from lime and urea application,” in 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Geneva: Intergovernmental Panel on Climate Change.

  • Google Scholar

• 61

  HughesT. P.KerryJ. T.Álvarez-NoriegaM.Álvarez-RomeroJ. G.AndersonK. D.BairdA. H.et al. (2017). Global warming and recurrent mass bleaching of corals. Nature543, 373–377. doi: 10.1038/nature21707

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  IEA (2021). Ammonia technology roadmap: Towards more sustainable nitrogen fertiliser production (Paris). Available online at: https://iea.blob.core.windows.net/assets/6ee41bb9-8e81-4b64-8701-2acc064ff6e4/AmmoniaTechnologyRoadmap.pdf (Accessed May 3, 2026).

  • Google Scholar

• 63

  IEA (2023). Net zero roadmap: A global pathway to keep the 1.5 °C goal in reach (Paris). Available online at: https://www.iea.org/reports/net-zero-roadmap-a-global-pathway-to-keep-the-15-c-goal-in-reach (Accessed November 3, 2025).

  • Google Scholar

• 64

  IMO (2008). MARPOL Annex VI — Regulations for the Prevention of Air Pollution from Ships (London: International Maritime Organization).

  • Google Scholar

• 65

  IMO (2015). International code of safety for ships using gases or other low-flashpoint fuels (IGF code). Resolution MSC.391(95) (London: International Maritime Organization). Available online at: https://www.imo.org/en/ourwork/safety/pages/igf-code.aspx (Accessed May 19, 2026).

  • Google Scholar

• 66

  IMO (2019). Resolution MEPC.323(74) Invitation to Member States to encourage voluntary cooperation between the port and shipping sectors to contribute to reducing GHG emissions from ships (London: International Maritime Organization). Available online at: https://www.imo.org/en/mediacentre/hottopics/pages/reducing-greenhouse-gas-emissions-from-ships.aspx (Accessed May 31, 2026).

  • Google Scholar

• 67

  IMO (2020a). “ Fourth IMO GHG study 2020,”(London: International Maritime Organization). Available online at: https://www.imo.org/en/ourwork/environment/pages/fourth-imo-greenhouse-gas-study-2020.aspx (Accessed May 26, 2026).

  • Google Scholar

• 68

  IMO (2020b). “ SOLAS Consolidated Edition 2020: Consolidated text of the International Convention for the Safety of Life at Sea 1974, and its Protocol of 1988,”(London: International Maritime Organization). Available online at: https://wwwcdn.imo.org/localresources/en/publications/Documents/Flyers/Flyers/IG110E.pdf (Accessed May 26, 2026).

  • Google Scholar

• 69

  IMO (2023). “ 2023 IMO strategy on reduction of GHG emissions from ships. Resolution MEPC.377(80),”(London: International Maritime Organization). Available online at: https://wwwcdn.imo.org/localresources/en/KnowledgeCentre/IndexofIMOResolutions/MEPCDocuments/MEPC.377(80).pdf (Accessed May 26, 2026).

  • Google Scholar

• 70

  IMO (2025). Report of the sub-committee on pollution prevention and response on PPR 13/17 (London: International Maritime Organization). Available online at: https://www.imo.org/en/mediacentre/meetingsummaries/pages/ppr-default.aspx (Accessed May 31, 2026).

  • Google Scholar

• 71

  IMO (2025a). Circular Letter No.5005 Draft revised MARPOL Annex VI (London: International Maritime Organization). Available online at: https://www.imo.org/en/mediacentre/hottopics/pages/cutting-ghg-emissions.aspx (Accessed May 31, 2026).

  • Google Scholar

• 72

  IMO (2025b). Interim guidelines for the safety of ships using ammonia as fuel. MSC.1/Circ.1687 (London: International Maritime Organization). Available online at: https://www.bimco.org/media/bxvcygg1/msc1-circ1687-interim-guidelines-for-the-safety-of-ships-using-ammonia-as-fuel-secretariat.pdf (Accessed May 31, 2026).

  • Google Scholar

• 73

  IRENA (2021). A pathway to decarbonise the shipping sector by 2050 (Abu Dhabi). Available online at: https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2021/Oct/IRENA_Decarbonising_Shipping_2021.pdf (Accessed May 31, 2026).

  • Google Scholar

• 74

  ISCC (2024). ISCC EU 205-1 renewable fuels of non-biological origin (RFNBO) and recycling carbon fuels (RCF) greenhouse gas emissions. Available online at: https://www.iscc-system.org/ (Accessed May 27, 2026).

  • Google Scholar

• 75

  JangH.Mujeeb-AhmedM. P.WangH.ParkC.HwangI.JeongB.et al. (2023). Regulatory gap analysis for risk assessment of ammonia-fuelled ships. Ocean Eng.287, 115751. doi: 10.1016/j.oceaneng.2023.115751

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 76

  JiS.ParkS.SeoY.ChoiM.LeeJ. (2025). Design and thermodynamic evaluation of onboard NH3 BOG re-liquefaction systems for ocean-going NH3 carriers. Int. J. Refrig.170, 399–411. doi: 10.1016/j.ijrefrig.2024.12.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 77

  JiangS.JinL.JinJ.IbánhezJ. S. P.WuY.ZhangJ. (2022). Exploring feedback mechanisms for nitrogen and organic carbon cycling in tropical coastal zones. Front. Mar. Sci.9. doi: 10.3389/fmars.2022.996655

  • CrossRef
  • Google Scholar

• 78

  JickellsT. D.BuitenhuisE.AltieriK.BakerA. R.CaponeD.DuceR. A.et al. (2017). A reevaluation of the magnitude and impacts of anthropogenic atmospheric nitrogen inputs on the ocean. Global Biogeochem. Cycles31, 289–305. doi: 10.1002/2016GB005586

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 79

  JonsonJ. E.JalkanenJ. P.JohanssonL.GaussM.Denier van der GonH. A. C. (2015). Model calculations of the effects of present and future emissions of air pollutants from shipping in the Baltic Sea and the North Sea. Atmos. Chem. Phys.15, 783–798. doi: 10.5194/acp-15-783-2015

  • CrossRef
  • Google Scholar

• 80

  KarlD. M.BidigareR. R.LetelierR. M. (2001). Long-term changes in plankton community structure and productivity in the North Pacific Subtropical Gyre: The domain shift hypothesis. Deep Sea Res. Part. II: Topical Stud. Oceanography48, 1449–1470. doi: 10.1016/S0967-0645(00)00149-1

  • CrossRef
  • Google Scholar

• 81

  KimT.-W.LeeK.NajjarR. G.JeongH.-D.JeongH. J. (2011). Increasing N abundance in the Northwestern Pacific Ocean due to atmospheric nitrogen deposition. Sci. (1979).334, 505–509. doi: 10.1126/science.1206583

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 82

  KirkelsA. F.LiuH.RomijnH. A.DurgaprasadS.PolinderH.GoudsmitM.et al. (2026). Batteries for sustainable shipping: Current status and potential roles. Renewable Sustain. Energy Rev.226, 116300. doi: 10.1016/j.rser.2025.116300

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 83

  KobayashiH.HayakawaA.SomarathneK. D. K. A.OkaforE. C. (2019). Science and technology of ammonia combustion. Proc. Combust. Inst.37, 109–133. doi: 10.1016/j.proci.2018.09.029

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 84

  Korean Register (2026). IMO news flash PPR 13 (Busan). Available online at: https://www.krs.co.kr/Common/Com_Popup/intradownload2.aspx?path=IMO+News+Flash(E)+PPR+13.pdf&filename=IMO+News+Flash(E)+PPR+13.pdf&physical=tech_infor (Accessed May 31, 2026).

  • Google Scholar

• 85

  KotzampasakisM.WooleyD. R. (2025). Mitigating carbon leakage in maritime transport. WMU J. Maritime Affairs. doi: 10.1007/s13437-025-00390-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 86

  KrishnamurthyA.MooreJ. K.MahowaldN.LuoC.DoneyS. C.LindsayK.et al. (2009). Impacts of increasing anthropogenic soluble iron and nitrogen deposition on ocean biogeochemistry. Global Biogeochem. Cycles23, GB3016. doi: 10.1029/2008GB003440

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 87

  KromM. D.HerutB.MantouraR. F. C. (2004). Nutrient budget for the Eastern Mediterranean: Implications for phosphorus limitation. Limnol. Oceanogr.49, 1582–1592. doi: 10.4319/lo.2004.49.5.1582

  • CrossRef
  • Google Scholar

• 88

  KuittinenN.KoponenP.VesalaH.LehtorantaK. (2024). Methane slip and other emissions from newbuild LNG engine under real-world operation of a state-of-the art cruise ship. Atmos. Environ. X23, 100285. doi: 10.1016/j.aeaoa.2024.100285

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 89

  LeeK.LiuX.VyawahareP.SunP.ElgowainyA.WangM. (2022). Techno-economic performances and life cycle greenhouse gas emissions of various ammonia production pathways including conventional, carbon-capturing, nuclear-powered, and renewable production. Green Chem.24, 4830–4844. doi: 10.1039/D2GC00843B

  • CrossRef
  • Google Scholar

• 90

  LejeusneC.ChevaldonnéP.Pergent-MartiniC.BoudouresqueC. F.PérezT. (2010). Climate change effects on a miniature ocean: The highly diverse, highly impacted Mediterranean Sea. Trends Ecol. Evol.25, 250–260. doi: 10.1016/j.tree.2009.10.009

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 91

  LelieveldJ.EvansJ. S.FnaisM.GiannadakiD.PozzerA. (2015). The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature525, 367–371. doi: 10.1038/nature15371

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 92

  MakinoT.WakayamaR.HattoriN.WaltherH.-P.MoserM.. (2025). “ Emission reduction technologies for full-scale two-stroke ammonia engines: Exhaust gas aftertreatment solutions & CCS,” in Cimac Congress 2025 ( MITSUI E&S, MAN Energy Solutions, Zurich). Available online at: https://papers2025.cimaccongress.com/pdf/CIMAC_paper_235.pdf (Accessed June 2, 2026).

  • Google Scholar

• 93

  MarkakiZ.OikonomouK.KocakM.KouvarakisG.ChaniotakiA.KubilayN.et al. (2003). Atmospheric deposition of inorganic phosphorus in the Levantine Basin, eastern Mediterranean: Spatial and temporal variability and its role in seawater productivity. Limnol. Oceanogr.48, 1557–1568. doi: 10.4319/lo.2003.48.4.1557

  • CrossRef
  • Google Scholar

• 94

  MMMCZCS (2023). Managing emissions from ammonia-fueled vessels. Available online at: https://www.zerocarbonshipping.com/publications/managing-emissions-from-ammonia-fueled-vessels/ (Accessed March 7, 2025).

  • Google Scholar

• 95

  MMMCZCSLloyd’s Register Maritime Decarbonization Hub (2025). Ammonia as fuel – competencies and training (Copenhagen). Available online at: https://www.zerocarbonshipping.com/publications/ammonia-as-fuel-competencies-and-training (Accessed May 31, 2026).

  • Google Scholar

• 96

  MohiteA. A.MathewE. (2025). Ship-generated waste management practices, challenges, and innovations onboard and at ports: A systematic review. Mar. Syst. Ocean Technol.20, 50. doi: 10.1007/s40868-025-00194-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 97

  MooreC. M.MillsM. M.ArrigoK. R.Berman-FrankI.BoppL.BoydP. W.et al. (2013). Processes and patterns of oceanic nutrient limitation. Nat. Geosci.6, 701–715. doi: 10.1038/ngeo1765

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 98

  MosierA.WassmannR.VerchotL.KingJ.PalmC. (2004). Methane and nitrogen oxide fluxes in tropical agricultural soils: Sources, sinks and mechanisms. Environ. Dev. Sustain.6, 11–49. doi: 10.1023/B:ENVI.0000003627.43162.ae

  • CrossRef
  • Google Scholar

• 99

  NaqviS. W. A.BangeH. W.FaríasL.MonteiroP. M. S.ScrantonM. I.ZhangJ. (2010). Marine hypoxia/anoxia as a source of CH ₄ and N ₂ O. Biogeosciences7, 2159–2190. doi: 10.5194/bg-7-2159-2010

  • CrossRef
  • Google Scholar

• 100

  NaqviS. W. A.NoronhaR. J. (1991). Nitrous oxide in the arabian sea. Deep Sea Res. Part. A. Oceanographic Res. Papers38, 871–890. doi: 10.1016/0198-0149(91)90023-9

  • CrossRef
  • Google Scholar

• 101

  OschliesA.BrandtP.StrammaL.SchmidtkoS. (2018). Drivers and mechanisms of ocean deoxygenation. Nat. Geosci.11, 467–473. doi: 10.1038/s41561-018-0152-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 102

  Pacific EnvironmentCSCEDF (2025). MEPC 83/7/23: REDUCTION OF GHG EMISSIONS FROM SHIPS Review of current literature on tank-to-wake nitrous oxide emissions from ammonia-fueled engines [Submitted to IMO MEPC Sub-Committee. Available online at: https://cleanshipping.org/wp-content/uploads/2025/02/MEPC-83-7-23-Review-of-current-literature-on-tank-to-wake-nitrous-oxide-emissions-fromammonia-fueled-en.-Pacific-Environment-CSC.pdf (Accessed May 31, 2026).

  • Google Scholar

• 103

  RabalaisN. N.DíazR. J.LevinL. A.TurnerR. E.GilbertD.ZhangJ. (2010). Dynamics and distribution of natural and human-caused hypoxia. Biogeosciences7, 585–619. doi: 10.5194/bg-7-585-2010

  • CrossRef
  • Google Scholar

• 104

  RavishankaraA. R.DanielJ. S.PortmannR. W. (2009). Nitrous oxide (N2O): The dominant ozone-depleting substance emitted in the 21st century. Sci. (1979).326, 123–125. doi: 10.1126/science.1176985

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 105

  RichardsonK.SteffenW.LuchtW.BendtsenJ.CornellS. E.DongesJ. F.et al. (2023). Earth beyond six of nine planetary boundaries. Sci. Adv.9 (37), eadh2458. doi: 10.1126/sciadv.adh2458

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 106

  RodrigueJ.-P. (2020). The Geography of Transport Systems. Fifth edition (Abingdon, Oxon ; New York, NY: Routledge). doi: 10.4324/9780429346323

  • CrossRef
  • Google Scholar

• 107

  RouwenhorstK. (2025). Emission performance of ammonia-fueled, four-stroke marine engines. Available online at: https://ammoniaenergy.org/articles/emission-performance-of-ammonia-fueled-four-stroke-marine-engines/ (Accessed May 31, 2026).

  • Google Scholar

• 108

  SandalowD.AinesR.FanZ.FriedmannJ.MccormickC.MerzA.-K.et al. (2022). Low-carbon ammonia roadmap. Available online at: https://www.icef.go.jp/wp-content/uploads/2024/02/icef2022_roadmap_Low-Carbon_Ammonia.pdf (Accessed March 8, 2025).

  • Google Scholar

• 109

  SchmidtJ.KonradsenF.KreutzfeldtK.-E. T.VemmelundS.NilssonM. (2025). Life cycle assessment of ammonia fuel (Denmark). Available online at: https://lca-net.com/files/Life-Cycle-Assessment-of-Ammonia-fuel-2025.pdf (Accessed April 15, 2025).

  • Google Scholar

• 110

  SchmidtA. L.WysmykJ. K. C.CraigS. E.LotzeH. K. (2012). Regional‐scale effects of eutrophication on ecosystem structure and services of seagrass beds. Limnol. Oceanogr.57, 1389–1402. doi: 10.4319/lo.2012.57.5.1389

  • CrossRef
  • Google Scholar

• 111

  ShortM. D.PetersG. M.PeirsonW. L.AshboltN. J. (2013). Marine nitrous oxide emissions: An unknown liability for the international water sector. Environ. Sci. Policy33, 209–221. doi: 10.1016/j.envsci.2013.06.003

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 112

  SignorD.CerriC. E. P.ConantR. (2013). N2O emissions due to nitrogen fertilizer applications in two regions of sugarcane cultivation in Brazil. Environ. Res. Lett.8, 15013. doi: 10.1088/1748-9326/8/1/015013

  • CrossRef
  • Google Scholar

• 113

  SmithC.HillA. K.Torrente-MurcianoL. (2020). Current and future role of Haber-Bosch ammonia in a carbon-free energy landscape. Energy Environ. Sci.13, 331–344. doi: 10.1039/c9ee02873k

  • CrossRef
  • Google Scholar

• 114

  SofievM.WinebrakeJ. J.JohanssonL.CarrE. W.PrankM.SoaresJ.et al. (2018). Cleaner fuels for ships provide public health benefits with climate tradeoffs. Nat. Commun.9, 406. doi: 10.1038/s41467-017-02774-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 115

  SteinL. Y.KlotzM. G. (2016). The nitrogen cycle. Curr. Biol.26, R94–R98. doi: 10.1016/j.cub.2015.12.021

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 116

  StevensC. J. (2019). Nitrogen in the environment. Sci. (1979).363, 578–580. doi: 10.1126/science.aav8215

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 117

  StieglmeierM.MooshammerM.KitzlerB.WanekW.Zechmeister-BoltensternS.RichterA.et al. (2014). Aerobic nitrous oxide production through N-nitrosating hybrid formation in ammonia-oxidizing archaea. ISME J.8, 1135–1146. doi: 10.1038/ismej.2013.220

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 118

  StoumposS.TheotokatosG.MavrelosC.BoulougourisE. (2020). Towards marine dual fuel engines digital twins—Integrated modelling of thermodynamic processes and control system functions. J. Mar. Sci. Eng.8, 200. doi: 10.3390/jmse8030200

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 119

  SunX.BuchananP. J.ZhangI. H.San RomanM.BabbinA. R.ZakemE. J. (2024). Ecological dynamics explain modular denitrification in the ocean. Proc. Natl. Acad. Sci.121(52), e2417421121. doi: 10.1073/pnas.2417421121

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 120

  SuntharalingamP.BuitenhuisE.Le QuéréC.DentenerF.NevisonC.ButlerJ. H.et al. (2012). Quantifying the impact of anthropogenic nitrogen deposition on oceanic nitrous oxide. Geophys. Res. Lett.39, L07605. doi: 10.1029/2011GL050778

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 121

  SuttonM. A.HowardC. M.ErismanJ. W.BillenG.BleekerA.GrennfeltP.et al. (2011). The European Nitrogen Assessment: Sources, Effects and Policy Perspectives (Cambridge: Cambridge University Press). doi: 10.1017/CBO9780511976988

  • CrossRef
  • Google Scholar

• 122

  TanE.YanX.DuM.ChenB.ChangY.ZouW.et al. (2025). Sedimentary nitrate respiration potentially offsets the climatic benefits from CO ₂ uptake by marginal seas. Geophys. Res. Lett.52, e2024GL112790. doi: 10.1029/2024GL112790

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 123

  TianH.PanN.ThompsonR. L.CanadellJ. G.SuntharalingamP.RegnierP.et al. (2024). Global nitrous oxide budget, (1980–2020). Earth Syst. Sci. Data16, 2543–2604. doi: 10.5194/essd-16-2543-2024

  • CrossRef
  • Google Scholar

• 124

  TianH.XuR.CanadellJ. G.ThompsonR. L.WiniwarterW.SuntharalingamP.et al. (2020). A comprehensive quantification of global nitrous oxide sources and sinks. Nature586, 248–256. doi: 10.1038/s41586-020-2780-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 125

  UeckerdtF.BauerC.DirnaichnerA.EverallJ.SacchiR.LudererG. (2021). Potential and risks of hydrogen-based e-fuels in climate change mitigation. Nat. Clim. Change11, 384–393. doi: 10.1038/s41558-021-01032-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 126

  UNEP-WCMCWorldFish CentreWorld Resources Institute (WRI)The Nature Conservancy (TNC) (2021). Global distribution of coral reefs. Available online at: https://resources.unep-wcmc.org/products/0613604367334836863f5c0c10e452bf (Accessed June 10, 2026).

  • Google Scholar

• 127

  U.S. EPA (2024). National emissions inventory (NEI) (Washington D.C: US Environmental Protection Agency). Available online at: https://www.epa.gov/air-emissions-inventories/national-emissions-inventory-nei (Accessed May 19, 2026).

  • Google Scholar

• 128

  U.S. EPA (2026). “ Toxic chemical release inventory reporting forms and instructions: section 313 of the emergency planning and community right-to-know act (Title III of the superfund amendments and reauthorization act of 1986),” (Washington D.C: US Environmental Protection Agency). Available online at: https://www.epa.gov/system/files/documents/2026-04/ry_2025_rfi_final_3.24.26.pdf.

  • Google Scholar

• 129

  Valera-MedinaA.XiaoH.Owen-JonesM.DavidW. I. F.BowenP. J. (2018). Ammonia for power. Prog. Energy Combust. Sci.69, 63–102. doi: 10.1016/j.pecs.2018.07.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 130

  Van DammeM.ClarisseL.WhitburnS.Hadji-LazaroJ.HurtmansD.ClerbauxC.et al. (2018). Industrial and agricultural ammonia point sources exposed. Nature564, 99–103. doi: 10.1038/s41586-018-0747-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 131

  Vega ThurberR. L.BurkepileD. E.FuchsC.ShantzA. A.McMindsR.ZaneveldJ. R. (2014). Chronic nutrient enrichment increases prevalence and severity of coral disease and bleaching. Glob. Change Biol.20, 544–554. doi: 10.1111/gcb.12450

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 132

  VoniatiG.DimaratosA.KoltsakisG.NtziachristosL. (2023). Ammonia as a marine fuel towards decarbonization: Emission control challenges. Sustainability15, 15565. doi: 10.3390/su152115565

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 133

  WanX. S.ShengH.-X.LiuL.ShenH.TangW.ZouW.et al. (2023). Particle-associated denitrification is the primary source of N2O in oxic coastal waters. Nat. Commun.14, 8280. doi: 10.1038/s41467-023-43997-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 134

  WangM.ElgowainyA.LeeU.BaekK.BalchandaniS.BenavidesP.et al. (2023). Greenhouse gases, regulated emissions, and energy use in technologies model ® (2023 Excel). doi: 10.11578/GREET-Excel-2023/dc.20230907.1

  • CrossRef
  • Google Scholar

• 135

  WangZ.LinL.YangM.XuY. (2016). The effect of future reduction in aerosol emissions on climate extremes in China. Clim. Dyn.47, 2885–2899. doi: 10.1007/s00382-016-3003-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 136

  WolframP.KyleP.ZhangX.GkantonasS.SmithS. (2022). Using ammonia as a shipping fuel could disturb the nitrogen cycle. Nat. Energy7, 1112–1114. doi: 10.1038/s41560-022-01124-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 137

  WongA. Y. H.SelinN. E.EasthamS. D.Mounaïm-RousselleC.ZhangY.AllroggenF. (2024). Climate and air quality impact of using ammonia as an alternative shipping fuel. Environ. Res. Lett.19, 084002. doi: 10.1088/1748-9326/ad5d07

  • CrossRef
  • Google Scholar

• 138

  WuB.WangY.WangD.FengY.JinS. (2023). Generation mechanism and emission characteristics of N2O and NO in ammonia-diesel dual-fuel engine. Energy284, 129291. doi: 10.1016/j.energy.2023.129291

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 139

  XuL.XuS.BaiX.-S.RepoJ. A.HautalaS.HyvönenJ. (2023). Performance and emission characteristics of an ammonia/diesel dual-fuel marine engine. Renewable Sustain. Energy Rev.185, 113631. doi: 10.1016/j.rser.2023.113631

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 140

  YangY.YangS.ParkS. (2026). Injection strategies for emission improvement in a direct-injection ammonia-diesel dual-fuel marine engine. Appl. Therm. Eng.285, 129256. doi: 10.1016/j.applthermaleng.2025.129256

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 141

  YeboahM.WijgerdeT.FoekemaE. M. (2026). Ammonia as a fuel for ships: A review of hazards and environmental dynamics during bunkering and shipping. J. Hazard. Mater. Adv.22, 101099. doi: 10.1016/j.hazadv.2026.101099

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 142

  YoshidaN.MorimotoH.HiranoM.KoikeI.MatsuoS.WadaE.et al. (1989). Nitrification rates and 15N abundances of N2O and NO3− in the western North Pacific. Nature342, 895–897. doi: 10.1038/342895a0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 143

  ZhangJ.GilbertD.GoodayA. J.LevinL.NaqviS. W. A.MiddelburgJ. J.et al. (2010). Natural and human-induced hypoxia and consequences for coastal areas: Synthesis and future development. Biogeosciences7, 1443–1467. doi: 10.5194/bg-7-1443-2010

  • CrossRef
  • Google Scholar

• 144

  ZhangI. H.SunX.JayakumarA.FortinS. G.WardB. B.BabbinA. R. (2023). Partitioning of the denitrification pathway and other nitrite metabolisms within global oxygen deficient zones. ISME Commun.3, 76. doi: 10.1038/s43705-023-00284-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 145

  ZhangY.ZanganehJ.MoghtaderiB. (2025). Advancement in neutral and negative emissions production of ammonia from carbonaceous sources: A review. Int. J. Hydrogen Energy109, 40–62. doi: 10.1016/j.ijhydene.2025.01.417

  • Pubmed Abstract
  • CrossRef
  • Google Scholar